1111111
111111

   
   

. ,1

1

11

   
 

11111
11

 

,111 1 "1
1 31111111111

01,1111.1l11.I

111111

.111, 111111111111
11,11 1111111
11,111111'11111111111'1111
11'1'111111111111'11'11111'1'11111111111'11.“

1 ,1,,,::1,111,1111111

'1
1111111111

, 1

' 1 :11 11'111 1,111, ,'11'11

1111 1'111

' 1

1 - 1,,,1 :1:,
1 1 , ,1

“1,111 .1111 11111 1 11,
111111..

111.111'111.,,}

11111
,',1,1,1,'.,

     
 

11,11

 

'11 1111111,,111111111.11111

.1111 ' 1' 11111111
111 1 1 1 11
11 ' .1". 11111111
' 11,11“1.7I11,1,1.:'1,1111,111'
.11 1 1 ,
111111,.11,11,1,,,,111 111,1 11.,11
'11111‘1'1' 11":1113'1111111'111'1
' 1' 1. '1 11111

,1111
.1‘

1,1111 11,11,
,1 1

1'"
1
.

.1. '1 111

  

111

1111'111'1111.
1,111,,,1111,11,

11,1 111111.1111

11.1" '1111 1'1'11

 

 

 

1,1111 '11111,1 ,,1,‘111

111111 11111111111111 11111111111111“ 111111111111111111111111111

W11

11111',111,'11
11,1111 1,111,-

if!) #31' 111$”

1111.11.

111111 111,111, 111.1

111111 11,

.11111111

111111111111

,1

 
    
   

,, 11111111
11111,,,,,,

    
    

,,,,1 111,,1 , 1111
,11111 | 1111') 1111
1111,111111'1'11

"11111111111111.1111
111111'1'1'111111,
11,11,111;1111 111,11"111,1:1111,11,1,1

11111111

‘I 1i‘.,
11111111111111» 111-11'
1 11 ,11 '11.,11111'111

1,111: 111111111
1,1,,1 ,11'11 1,111 ,,,, ,1111'

11 1, ,.,,,

1'11 111 1'

,111,',111,:111' 11
11 1111 1' 11‘1'1

111
.11 1111 11111

11W-
1

111111111 1111

,1 '11 11 11
1., 1, 1 1111 1,,1,1,,1,1,, 11 111,, 11111111

111111 1111111111

‘wa ————.~

”511,111., 1111171 ,1111,,,, “F111

,,,":u,1,§',,, 1,,)“ 11,11,11111'1111111111'11 111111,
1,111., “£1 ,1, ,‘1‘11911'1,

111111, ........ "1:1
.1111 1111

'- 1 11111.
111111.11 1111111 111111,,111111

1,1 ,
1 1.1 1111111.

'1l|"1,1

11,1 1

11.11

111,111 1

11', .1, ,
1111 '1111 ,1111
11111 111111' 11 1111111

1111

111 1111111,

11' 1
111111

 
 
   
  

1111

1,,1,11 11111,1,,1
1 ,1,l,, ,11 11:,,,,,11 1,1::1.11111
'11, 1,11'1111111'1111
111111-11 ,11111,,,

1.1,, 111
111,1'11

11111, 1 1

 

111111111111 111

1,1,1111, ,,, 1,1 1,,,
1,l1111 1,1 1,1,1" ,11
11,11 ,111,.1,,1,',,111.|11,11,11,,,1,1,1,

111 1,11 11 1
1 11 111111
11111111,1111111111'1111 1111 111111

111

  
  

11

1111111111 111111111

1.11.17."1

1,,
11411111111111.1111 1’11
1'171'111', ' 11 1111

11111111111111

 
 
    
 
 
  
 
     
     
   
   
 
 
 

111

    
    
   

111111111

11'
1

  
 

1,1'
1111,,
111

111

|

   

 
 
 
  
 
 

111
'1111111 1111

 
  
   
 

11,11.

1,111,111 ,‘11191:

 

11'

11:1'11, 1

111 11,
111111181 111'1 1'11

     
 
   
 
  
 
   

,1111111
111111

111111,,

111’

.,. 177:7?

111 '1

. (‘1! ‘1’ "1.
. ' 11|1

      

   
 
  
 
  
 
   

  
  
 
  
   
 
 
 
 
 
   

1111
111 11'
I

Efl'lin-CLLIQ'T'xI v.’ 1 1"2?“
. -‘ ' ‘-
7‘ a " .-" Q I
«ruccm " ' ‘ - 1‘
.

Pvlicliigze. r: 5mm
University

 
     

This is to certify that the

thesis entitled

CATALYSIS BY TITANOCENE DERIVATIVES

presented by

John D. Bay

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Chemi stry

 

 

fiflmaédMfiJMw

Majé/r professor

Date .. I X 9 X

0-7 639

 

0mm}: FINES ARE 25¢ PER DAY .
PER ITEM

Return to book drop to remove
this checkout from your record.

 

 

 

 

CATALYSIS BY TITANOCENE DERIVATIVES

By

John D. Ray

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirement

for the degree of
DOCTOR OF PHILOSOPHY
Department of Chemistry

1978

ABSTRACT

CATALYSIS BY TITANOCENE DERIVATIVES

By
John D. Ray

Titanocene dichloride has been used for Ziegler-

l 2

Natta polymerization and nitrogen fixation experiments
as a precursor to the active catalyst. A polymer attached
titanocene dichloride can, after reduction with BuLi,
catalyze hydrogenation of olefins, hydroformylation, and
diene isomerization.3’u Titanocene is a reactive, co-
ordinatively unsaturated species that is predicted to have
carbene-like activity.5 Singlet carbenes undergo facile
insertion into C-H sigma bonds, addition to pi bonds of
alkenes and alkynes, and addition to pi acceptors like CO
and N2.6 Besides accounting for the reactions of titano-
cene with other compounds, the titanocene carbene-like
behavior leads to rearrangement within the molecule it-
self. Thus, titanocene deactivates to an inactive ful-
valenide bridged dimer, [(CSH5)TiH]2(ClOH8).7 Attach-
ment of titanocene dichloride to 20% crosslinked poly—

(divinylbenzene-styrene) prevents deactivation by

John D. Ray

dimerization.

The research reported here attempts to further identify
the polymer attached titanocene species by (1) an esr
study of the titanocene derivatives and comparison to the
polymer attached compounds, (2) a comparison of hydrogena-
tion activity of homogeneous titanocene derivatives compared
to the polymer attached compounds, (3) characterization
of the active species by its reactions and by preparation
using alternative routes. In addition, the activity of
the reduced polymer attached species toward nitrogen was
of interest because the homogeneous titanocene compounds
that have been used for nitrogen fixation. It was
thought that reaction of the carbene-like titanocene with
nitrogen could lead to formation of ammonia if the di-
merization to inactive "titanocene" could be prevented.
Comparison reactions of titanocene species with the
reduced polymer species led to the discovery of catalytic
activity by some titanocene derivatives, especially the
monohydride dimer (Cp2T1H)2.

A convenient method for the preparation of the mono-
hydride dimer, (Cp2TiH)2, in high yields in a hydro-
carbon solvent has now been developed. Titanocene di-
chloride may be reduced in hexane under one atmosphere
of hydrogen to the purple monohydride dimer by use of
two equivalents of butyl lithium (BuLi). The purple

hydride dimer is an active catalyst for hydrogenations

John D. Ray

of olefins and isomerization of dienes. Hydrogenation
rates obtained with the purple hydride and l-hexene are
approximately 30 times faster than the reduced titano-
cene dichloride beads and hundreds of times faster than
the homogeneous Cp2TiCl2 reduced at room temperature.
Isomerization of allylbenzene, 1,5-cyclooctadiene, and
l-octene occurs readily even at 0°. The purple hydride
also reacts with N2 and the complex formed may be hydroly—
ized to yield greater than 1 mmol NH3/mmol Ti. The cata-
lytic activity of (szTiH)2 is compared to reduced polymer
supported titanocene dichloride species, Cp2TiH2',
Cp2TiCl, Cp2Ti(CH2Ph), and THF or PPh3 derivatives.

The identity of the reduced polymer supported titano-
cene species is still unknown. ESR spectra of the species
are consistent with either a Ti (II) or Ti (III) compound
on the polymer. Treatment of the reduced polymer attach-
ed species indicates the presence of a hydride. The di-
hydride anion, Cp2T1H2-, species can be ruled out on the
basis of esr spectra and catalytic activity. Thus, the
polymer attached species is probably the titanocene
analog ((:)—Cp2Ti), the hydride derivative from ring
hydrogen abstraction, or the titanocene monohydride deri-
vative.

No nitrogen could be fixed for either the reduced

polymer supported Cp2T1C12 or for the polymer supported

Cp2Ti(CH2Ph). Neither high pressure nor lower temperatures

 

John D. Ray

were effective for improving the nitrogen fixation.

A dimeric titanocene species is apparently required for
binding the dinitrogen. Such dimeric binding cannot
occur with the polymer supported titanocene compounds and

a dinitrogen complex does not form.

BIBLIOGRAPHY

1. F. A. Eisch and R. B. King, "Organometallic Synthesis",
Vol. 1, Academic Press, New York, NY, 1965.

2. P. C. Wailes, R. S. P. Coutts, and H. Weigold, "Organo—
metallic Chemistry of Ti, Zr, and Hf", Academic
Press, New York, NY, 197“.

3. W. D. Bonds, Jr., C. H. Brubaker, Jr., E. S. Chandrese-
karan, C. Gibbons, R. H. Grubbs, and L. C. Kroll, Jr., J
Am. Chem. Soc., 21, 2128 (I975).

 

A. C. P. Lau, Thesis, Michigan State University, 1977.

5. H. H. Brintzinger and L. S. Bartell, J. Am. Chem. Soc.,
2, 1105 (1970).

6. W. Kirmse, "Carbene Chemistry", Academic Press, New
York, NY, 196a.

7. J. E. Bercaw, R. H. Marvich, L. G. Bell, and H. H.
Brintzinger, J, Am. Chem. Soc., 2A, 1219 (1972).

 

ACKNOWLEDGMENT

The author wishes to express his gratitude to Pro-
fessors Carl H. Brubaker and Robert H. Grubbs for their
guidance and support during this research.

Thanks also to Linda, who provided encouragement,

happiness, and perspective during my graduate years.

11

TABLE OF CONTENTS

Page
LIST OF TABLES. . . . . . . . . . . . . . . . . vi
LIST OF FIGURES . . . . . . . . . . . . . . . . viii
INTRODUCTION. . . . . . . . . . . . . . . . . . 1
EXPERIMENTAL. . . . . . . . . . . . . . . . . . ll
General . . . . . . . . . . . . . . . . . . ll
Chloromethylation of Copolymer Beads. . . . 13
Preparation of Cyclopentadienyl-
substituted Copolymer . . . . . . . . . . 15
Preparation of Polymer Attached
Titanocene Dichloride . . . . . . . . . . . 15
Preparation of CpTiCl3. . . . . . . . . . . 16
Preparation of Polymer Attached
Dimethyl Titanocene . . . . . . . . . . . . 17
Preparation of Titanocene Mono-
chloride. . . . . . . . . . . . . . . . . . 17
Preparation of Benzyl Titanocene. . . . . . l7
Nitrogen Fixation Using Benzyl
Titanocene. . . . . . . . . . . . . . . . . 18
Preparation of Polymer Supported
Titanocene Monochloride . . . . . . . . . . 19
Preparation of Polymer Attached
Benzyl Titanocene . . . . . . . . . . . . . 20

Treatment of Polymer Attached
Benzyl Titanocene with Nitrogen . . . . . . 20

Preparation of Titanocene Monohydride
Dimer ... . . . . . . . . . . . . . . . . . 21

Reactions of the Titanocene Mono-
hydride Dimer . . . . . . . . . . . . . . . 22

Preparation of Titanocene Mono-
hydride from the Dimethyl Deriva-
tive. O I O O O O O O O O O O O O I O O O 0 26

Hydrogenation Using Titanocene
Monohydride Dimer . . . . . . . . . . . . . 26

iii

Page

Hydrogenation Using Titanocene

Dihydride Anion . . . . . . . . . . . . . . 27

Hydrogenation with Reduced Homo-

geneous CpZTiCl2. . . . . . . . . . . . . . 28

Reduction of ® -Cp2TiC12 and

Treatment with DCl. . . . . . . . . . . . . 28

Nitrogen Fixation by Use of

Homogeneous Cp2T1012. . . . . . . . . . . . 30

Nitrogen Fixation by Use of

Polymer Supported Cp2TiC12. . . . . . . . . 31
RESULTS AND DISCUSSION. . . . . . . . . . . . . 32

Hydrogenation Studies . . . . . . . . . . . 39

Comparison of the Beads with

(Cp2T1H2)‘. . . . . . . . . . . . . . . A2
Preparation of Titanocene Mono—

hydride Dimer . . . . . . . . . . . . . Uh
Hydrogenation by (Cp2TiH)2. . . . . . . A7

Effect of Temperature on
Rates . . . . . . . . . . . . . . . . . A8

Comparison of Hydrogenation

Rates with Bead Species . . . . . . . . 6
Comparison of Catalyst Rates. . . . . . 59
Reactions of (Cp2TiH)2....... ...... . . . . 69

Catalytic Isomerization of
Allylbenzene. . . . . . . . . . . . . . 69

Isomerization of 1,5-cyclo-
octadiene . . . . . . . . . . . . . . . 69

Isomerization of l—octene . . . . . . . 7O
Nitrogen Fixation Experiments . . . . . . . 72

Volpin and Shur High Pressure
Method. . . . . . . . . . . . . . . . . 72

Fixation by Cp2Ti(arene)
Compounds . . . . . . . . . . . . . . . 7U
Fixation of N2 by (Cp2TiH)2 . . . . . . 75

D01 Treatment of Polymer Supported
Titanocene Species. . . . . . . . . . . . . 78

iv

ESR of the Copolymer Bead
Species . . . . . . .

APPENDIX A. Attempted Preparation of
Polymer Supported Molybdenocene Di-
chloride Through the ® -CpMoC1u
Intermediate. . . . . . . . . . .
Discussion. . .

Experimental.

APPENDIX B. Preparation and Attempts to
Reduce Polymer Attached Molybdenocene
Dichloride. . . . . . . . . .
Discussion. . . . . . . . . . . . . .

Experimental. . . . . . . .

APPENDIX C. Preparation of Polymer
Attached Group VI CpM(CO)3H Compounds

Discussion. . . . . . .

Experimental. . . . . . . . . .
APPENDIX D. Preparation of Polymer

Attached Group IV Arene Metal Tri-

carbonyl Compounds. . . . . . . . .

Discussion. . . . . . . . . . . . . .

Experimental. . . . . . . . . . . . .

BIBLIOGRAPHY. . . . . . . . . . .

Page

8A

10A
10A

110

113
113

120

12“
12A
127

129
129
131

133

Table

10

LIST OF TABLES

Page
Hydrogenation rates compared -
polymer attached species and
(Cp2TiH)2. . . . . . . . . . . . . . . . . A9
Hydrogenation rates at 25°C for
cyclohexene. . . . . . . . . . . . . . . . 50
Effect of olefin concentration
on hydrogenation rates . . . . . . . . . . 51
Effect of Ti concentration on
hydrogenation rates. . . . . . . . . . . . 52
Effect of incremental olefin
additions on hydrogenation by
(Cp2TiH)2. . . . . . . . . . . . . . . . . 55
Effect of THF or PPh3 on hydrogena-
tion rates . . . . . . . . . . . . . . . . 57
Hydrogenation with beads treated
with 2 equivalents of BuLi . . . . . . . . 58

Hydrogenation with ® 4132111012

plus homogeneous Cp2TiC12. . . . . . . . . 59
Nitrogen fixation results - Volpin

and Shur method. . . . . . . . . . . . . . 73
Nitrogen fixation yield by use of

(Cp2TiH)2. . . . . . . . . . . . . . . . . 77

vi

Table Page

11 Composition of liberated gas from

DCl treated reduced®-Cp2TiC12. . . . . . 79
12 Summary of esr values for titano-

cene derivatives . . . . . . . . . . . . . 99
13 ESR spectral values for MoClS. . . . . . . 109

1A Far—ir spectral data of Cp2MoC12

compounds. . . . . . . . . . . . . . . . . 116
15 Summary of attempted reactions

with®-Cp2MoCl2. . . . . . . . . . . . .118
16 Summary of attempted reactions

with®-Cp2MoH2. . . . . . . . . . . . . .119

vii

LIST OF FIGURES

Figure Page

1 Septum capped reaction flask with

esr side tube. . . . . . . . . . . . . . . 12
2 Reactions of titanocene. . . . . . . . . . 38
3 Reactions of titanocene mono-

hydride dimer. . . . . . . . . . . . . . . 38
A Graph of hydrogenation temperature

versus rate. . . . . . . . . . . . . . . . 53
5 Relative hydrogenation rates for

bead catalyst and (Cp2TiH)2. . . . . . . . 61
6 Relative hydrogenation rates after

THF or PPh3 addition . . . . . . . . . . . 62
7 Graph of hydrogenation rates versus

(CpTiH)2 concentration . . . . . . . . . . 63
8 Comparison of hydrogenation rates

for titanocene catalysts . . . . . . . . . 67

9 Mechanisms for hydrogenation by
titanocene derivatives . . . . . . . . . . 68
10 Graph of percent hydrogen isotope
versus reaction temperature. . . . . . . . 8O
11 ESR spectrum of BuLi reduced
®-Cp2T1012................85
12 ESR spectrum of®-Cp2TiC1 . . . . . . . . 87

viii

Figure Page

13 Frozen solution esr spectrum

of Cp2TiC12. . . . . . . . . . . . . . . . 88
1A ESR spectrum of®-Cp2Ti(CH2Ph). . . . . . 90
15 ESR spectra of Cp2Ti(Bu)E. . . . . . . . . 92
16 ESR spectrum of (Cp2Ti(Bu)H' . . . . . . . 93
17 ESR spectrum of (Cp2TiH2)- . . . . . . . . 9A
18 ESR spectrum of (Cp2TiH2)' with Mg

Hyperfine splitting. . . . . . . . . . . . 95
19 Frozen solution esr spectrum of

(Cp2TiH2)- . . . . . . . . . . . . . . . . 97
20 Effect on an esr quintruplet of

freezing the solution. . . . . . . . . . . 98
21 ESR spectrum of Cp2Ti(PPh3)H . . . . . . . 101
22 ESR spectrum of (Cp2TiH)2 in

THF. . . . . . . . . . . . . . . . . . . . 103
23 Preparation scheme for

®-Cp2M0012................1O6
2A Alternative preparation scheme for

®-Cp2MoC12................106
25 Scheme for the Preparation of Polymer

Supported Molybdenocene di-

chloride . . . . . . . . . . . . . . . . . 11A
26 ESR Spectrum of Beads obtained from

P -Cp2MoC12 synethsis. . . . . . . . . . . 115

ix

INTRODUCTION

Transition metal complexes that are catalysts and are
bonded to an insoluble polymer support lie somewhere
between the usual classifications of heterogeneous or
homogeneous species. Since the polymers are insoluble
and filterable, the overall polymer as a catalyst must
be considered to be a heterogeneous catalyst. Our usual
understanding of a heterogeneous catalyst is a solid con-
taining a reactive surface and may have several different
active sites. The metal site on a homogeneous catalyst,
however, has a defined steric and electronic environment.
Only one type of identical active center is present, and
so a high degree of substrate selectivity may occur.
Organometallic complexes that behave as homogeneous catalysts
may be made insoluble by attachment of a bulky ligand to
the metal. The reactive site is changed very little when
a bulky polymer is a side chain on the ligand. Thus the
insoluble catalyst formed is more like a homogeneous
catalyst in reactivity and selectivity.

Several authors have offered new definitions to avoid
the homogeneous-neterogeneous classifications. Homogeneous
catalysts attached to insoluble polymer supports have thus;

been referred to as "heterogenized" homogeneous catalysts,l

2

"solid-phase" synthesis, hybrid phase catalysts,3 or

homoactive heterogeneous catalysts."l The latter describes

the system well. Homoactive catalysts have only one type
of active center and the words homogeneous and hetero-
geneous refer only to the number of phases present.

Homogeneous catalysts have found only limited use in
industrial chemistry, partly because of the difficulty
in their separation from the reaction products. Limited
temperature ranges of activity, loss of catalytic activity,
and need for batch rather than flow systems are drawbacks.
Polymer attachment of a homogeneous catalyst has the po-
tential of significantly improving their industrial appli-
cability. Insoluble bead polymers allow separation of the
catalyst from the liquid phase components by filtration.
Columns of supported catalyst could easily be used for
flow systems. The homoactive catalyst retains the
specificity, efficiency, and controllability inherent in
homogeneous catalysis. The support, furthermore, allows
usecfi‘compounds that might dimerize or otherwise be de-
activated by interaction of metal sites.

There are some disadvantages to the polymer supported
catalyst systems. Identification of the active catalyst
"intermediate" is more difficult because of the surrounding
polymer. In some cases reactions may be diffusion limited
and only limited temperatures and pressures can be used.
More than one species may be unintentionally attached to
the polymer which makes separation nearly impossible.

The effectiveness and selectivity then would be reduced.

Indications are that such is the case during the synthesis
procedures, as each reaction is not 100% effective. Some
catalysts are known to be leached from the polymer leading
to a loss in activity and of the metal complex.5

Several methods of attaching a catalyst to a polymer
have been developed. Ion exchange resins have been used

with charged or polar species.6

Adsorption into non-
functionalized polymer has been used to advantage with
A1C13.7 Generally, covalent chemical bonding as pendant
functional groups to the polymer is viewed as stronger,
more desirable attachment. In some cases monomer units
of the catalyst can be copolymerized so that the catalytic
moieties are part of the polymer backbone.8
With covalent linkage to the polymer, one may consider
the supported catalyst as an insoluble polymer framework
with pendant catalyst projecting into the solvent and
in a sense solvated. An alternate view would be to
consider dilute solvated catalyst complexes as having
attached a bulky insoluble ligand. Although the polymer
may limit diffusion and control the size of incoming sub-
strate, it neither appears to effect the electronic en-
vironment of the metal nor its reactivity significantly.
A number of metal complexes have now been attached
to organic polymer supports.9 Haag and Whitehurst6
were pioneers in the field of polymer attachment of

homogeneous complexes for industrial use. They attached

2+

[Pt(NH3)u] to sulfonate resins and used this material

to catalyze the reaction of allyl chloride and carbon mon-

10 these authors disclosed a

oxide. In another patent,
method to coordinate rhodium chloride to polystyrene polymer
with pendant phosphine groups. The resultant polymer metal
complex was an active catalyst for hydrogenation.

Grubbs and Kro11ll have attached Wilkinson's catalyst,
tristriphenylphosphine rhodium (I) chloride, to 2% cross-
linked polystyrene divinylbenzene resin through a tri—
phenylphosphine function on the polymer. The polymer was
an active hydrogenation catalyst that could be "reused"
repeatedly without loss of activity. In addition the
polymer catalyst was capable of substrate size selectivity
in the rate of hydrogenation of olefin molecules.

Pittman,12

§£_§J, have synthesized a variety of polymer
bound transition metal (Mo, Fe, Mn, Ni, Co, Rh) carbonyl
catalysts. The catalyst center could be bound to linear
or crosslinked polystyrene by coordination to a phosphine
or by copolymerization with an organometallic monomer
containing a vinyl group. They found the metal resins
to be useful hydroformylation, oligomerization, isomeriza-
tion catalysts.

Grubbs, Brubaker, and coworkers13 were able to attach
titanocene dichloride to a polystyrene-divinylbenzene co-

polymer. On reducing with butyl lithium (BuLi) a catalyst

was obtained that had a hydrogenation efficiency several

times greater than that of the homogeneous species under
similar conditions. Like the polymer attached Wilkinson's
catalyst, the titanocene resin could be reused with only
slight loss in activity. In addition, the attached titano—
cene dichloride could be regenerated from the active catalyst
by treatment with HCl. The polymer attached titanocene
catalyst has been shown to also promote isomerization,
oligomerization, and hydroformylation.lu
Organometallic compounds containing a cyclopentadienyl

ligand are attached to macroreticular polystyrene-divinyl-

benzene copolymer beads by the following reaction scheme:

CICH2_O-C2H5

SnCl4 a

Polystyrene H2C|

 

CH3Li

‘ Tl-lF
./CH2

 

 

Treatment of the cyclopentadienyl substituted resin

with a transition metal compound leads to attachment of

the metal to the polymer through pi bonding to the cyclo-
pentadienyl ring. The polymer bound titanocene is prepared

by treatment of (A) with cyclopentadienyl titanium tri-

chloride.

CH
“or
CH Ti H

2 m @é‘.
@f

2
/Cl
\Cl

An alternate scheme for attaching metallocenes to
the copolymer is based on the observation by J. Lee that
(hS-CSH5)2Ti(hl—C5H5)Cl undergoes intermolecular sigma-
pi ring exchange. This ring exchange reaction has now
been used to attach a variety of metals15 (Ti, Zr, Hf,

Hb, Mo, W).

CH2 CH

‘iiuf‘ ‘ny
[§;{£’m;——-> [:§Zfitcr—fi5L—9
@ < @l %M

A A
'88) $2

or” o“?

A A A
c

<2!

4—

To avoid the use of Chloromethylated polymer that
requires handling the cancer suspect agent chloromethyl
ethyl ether,16 a method developed by L. Krollll can be
used to attach titanocene dichloride. The amount of ti-
tanium attached to the polymer is generally less than
when the other methods are used. No difference in activity
is noted due to the slightly different linkage of the

titanocene dichloride to the polymer backbone.

 

Since polymer attachment of the titanocene species by

the methods above lead to metal center isolation because
of the rigid support, dimerization of the reduced titano-

cene dichloride species is prevented. The homogeneous

H/Hs \‘i

so /‘ 2d

titanocene undergoes a cyclopentadienyl sigma-hydrogen

abstraction followed by dimerization to form a fulvalenide

5_ 17
bridged hydride,[(h C5H5)T1H]2(C1008). This is the

ea or,
a

. H
Ti '-/

1-0 3

IV

 

 

inactive green "titanocene" prepared by Watt and Drummond.18

The polymer supported titanocene species is prevented from
undergoing this type of deactivation, that results in an
increase in activity for olefin hydrogenation. Site
isolation is further supported by observation of the rate
of hydrogenation of l—hexene compared to titanium loading
on the polymer.1u
Because of the difficulty in identifying compounds
with the polymer, the exact nature of the titanocene
species has not been specified.15 In addition to titano-
cene itself, several titanocene hydrides or alkyl complexes
are also possibilities for being the active species on
the polymer. Observation of an esr signal at g = 1.99
is consistent with the active species being Ti(II)<n~
GH.(III)but without hyperfine splitting further identifica-
tion is difficult. That treatment of the reduced species

regenerates polymer attached titanocene dichloride indicates

10

that the bis—n-cyclopentadienyl integrity is retained
throughout the reduction procedure.
The research work reported here attempts to further

identify the polymer attached titanocene Species by:

(1) An esr study of the titanocene derivatives and
a comparison of those spectra to those for the
polymer attached compounds;

(2) A comparison of hydrogenation activity of homo-
geneous titanocene derivatives compared to the
polymer attached compounds;

(3) Characterization of the active species by its
reactions and by preparation using alternate

routes.

In addition, the activity of the reduced polymer attached
species toward nitrogen was of interest because of the
homogeneous titanocene compounds that have been used for
nitrogen fixation. It was thought that reaction of the
carbene-like titanocene with nitrogen could lead to forma-
tion of ammonia if the dimerization to inactive "titano-
cene" could be prevented. Comparison of reactions of
titanocene species with the reduced polymer species led to
the discovery of catalytic activity by some other titano—
cene derivatives, especially the monohydride dimer,

(szTiH)2.

EXPERIMENTAL

Manipulations involving air—sensitive materials were
performed under argon or nitrogen in Schlink-type (air—
less ware) vessels. Transfers of air sensitive reagents
and reaction solvents were often made by using glass
syringes. For small amounts of polymer supported compounds,
50 m1 septum capped erlenmeyer flasks were used. An
inert-gas vacuum manifold fitted with syringe needles
held the flasks during reaction. To transfer solids an
argon filled glove box was used. Many preparations of
moderately air-sensitive compounds were carried out in 100
m1 round bottom flasks with stopcock sidearms fitted with
rubber septums. Preparations of compounds for esr experi-
ments were often carried out in septum capped 50 m1 flasks
with esr side-tubes (Figure 1).

Spectra were measured with samples prepared by grind-
ing the polystyrene beads in a Wig-l-Bug under inert gas
and then were mulled in dry Nujol. Far infrared spectra
in the 100-500 cm"1 region were obtained by use of a
Digilab model FTS-l6 Fourier transform spectrometer.
Spectra were recorded with the sample in dry nitrogen
atmosphere and mounted between polyethylene plates.
Infrared spectra were obtained by means of a Perkin Elmer
"57 or a 237B grating spectrophotometer in Nujol mulls.

The esr spectra were recorded by use of a Varian E-A

11

12

I)

 

 

 

 

 

 

Figure l. Septum capped reaction flask with esr side
tube.

13

esr spectrometer and the instrument calibrated against
or by use of a pitch sample. Mass spectra were recorded
by use of Hatachi-Perkin Elmer RMU-6 mass spectrometer.

Reagent grade diethyl ether, tetrahydrofuran (THF),
hexane, benzene, and toluene were dried by being refluxed
under argon with sodium-benzophenone or lithium aluminum
hydride, and distilled under argon immediately before use.
Argon and nitrogen gases were purified before use by passing
through a column of activated copper catalyst (BTS cata—
lyst) and subsequently through Aquasorb (phosphorous
pentoxide on an inert support). Titanocene dichloride
obtained from Alfa Products was extracted by use of a
Soxhlet extractor and recrystallized from toluene.

The 20% crosslinked (600 A pore size) and 2% cross-
linked macroreticular polystyrene-divinylbenzene copolymer
beads were gifts from the Dow Chemical Company. Before
use the polymer beads were sieved to A0-60 mesh and washed
to remove impurities. Organolithium reagents were ob-
tained from Alfa Products or Aldrich Chemical Company.
Chloromethyl ethyl ether was obtained from Aldrich and

distilled carefully before use.16

Chloromethylationggf,Copglymer Beadsl9

A batch of 20% crosslinked copolymer beads was
sieved to obtain a 32—35 mesh size. These beads were

then washed with 10% HCl, 10% NaOH, 1:1 H2O:CH3OH,

1A

CH OH, 1:1 CH3OH:CH2Cl2, CH2C12, and benzene to remove

3
impurities, as recommended by Pittman.2O They were then
vacuum dried. Following the Chloromethylation method of

21 50 g of washed and dried copolymer beads

Pepper, 2E ml.
were placed into a one liter three-neck flask fitted with

a drying tube, addition funnel, and overhead stirrer.

After the addition of 300 mL of freshly distilled chloro-
methyl ethyl ether,16 the mixture was stirred for 15 hrs.
The flask was cooled in an ice bath and then a cold solu-
tion of 8.1 mL of SnClu in 100 mL of chloromethyl ethyl
ether was introduced dropwise. The beads and solution
turned a light pink as the SnClu was added. After addi-
tion of the SnClu solution was completed, the reaction mix-
ture was allowed to warm to room temperature and stirred
for 13 hrs. The chloromethyl ethyl ether solution was
removed by suction through a fritted dispersion tube.

The beads were washed with three 300 mL portions of 50%
aqueous dioxane, two portions aqueous dioxane containing
10% HCl, and finally with dry dioxane until the washings
were chloride free as determined with a test AgNO3 solu-
tion. The beads were further washed with two 200 mL
portions of benzene and then dried by evacuating for two
days a flask placed in a warm water bath. Chloride
analysis19’22 of the white beads indicated 0.79 mmol

Cl/g of copolymer, which is equivalent to 8% Chloromethyla-

tion of the styrene rings.

15

Preparation of Cyclopentadienyl-Substituted Copolymer

The chloromethylated copolymer beads were suspended
in 100 mL of THF in a one liter three-neck flask fitted
with a gas inlet tube, a septum capped Joint, and an over-
head stirrer. A three-fold excess of freshly prepared
sodium cyclopentadienide (Nan) in THF was added using a
double tipped needle. The mixture was stirred at room
temperature for 2 days before removing the excess Nan
by filtration. The beads were washed with 3x 100 mL THF,

once with 1:1 C OH:THF, and finally with 3x 80 mL

2H5
benzene. The yellow beads were then dried by evacuation
of the flask. Chloride analysis indicates 0.06 mmol/g

of polymer, which represents 0.73 mmol CpH/g of polymer.

Preparation of Polymer Attached Titanocene Dichloride

The 50 g batch of cyclopentadiene substituted copolymer
beads was placed in a three-neck flask equipped as before.
The beads were first suspended in 100 mL dry THF and then
a three-fold excess of methyl lithium (or butyl lithium)
in THF added. This mixture was stirred in the dark for
two days at room temperature under an atmosphere of
argon. Excess methyl lithium was removed by using a
syringe. After being washed three times with THF and
twice with benzene, the beads were suspended in 100 mL

of benzene. A two-fold excess of CpTiCl3 dissolved in

l6

benzene was added to the suspension of beads, and the
mixture stirred another three days in the dark. Excess
CpTiCl3 was removed by first being washed 3x 80 mL with
benzene and then by soxhlet extraction with benzene. The
polymer supported titanocene dichloride beads were pink
or light red after being dried $2.YESEQ-

Analysis: 0.2A mmol Ti/g polymer

0.50 mmol Cl/g polymer

ClzTi ratio 2.08

far-ir- polymer attached 220, 255, 310, 386, A10 Cm-l
homogeneous 207, 2A6, 307, 360, A03 Cm'l

diamagnetic

23

Preparation of CpTiCl3

A mixture of 12 g (0.0A8 mole) Cp2TiC12, 1A.6 mL
(0.136 mole) TiClu, and 90 mL of xylene was refluxed in
a flask for 2.5 hours under a nitrogen atmosphere. After
being cooled, the solution was filtered and dirty yellow
crystals were obtained. The crude product and granular
charcoal were placed in a glass soxhlet thimble and the
CpT1013 was extracted with dry benzene. Yellow crystals
were collected by filtration of the extract and the product
recrystallized from fresh benzene.

Bright yellow crystals

mp.p. 208-210°

far-1r - 290, 332, Aled, A58 Cm‘l

17

Preparation of Polymer Attached Cp2Ti(CH3l2ii

Polymer attached titanocene dichloride beads, 1.1
g (0.25 mmol), and 10 mL of ether were stirred together
in a 50 mL septum capped flask. After the mixture was
cooled to -30° under argon, 0.A8 mmol of CH3Li in ether
was added slowly and the mixture allowed to stir for 2
days. The yellow beads were extracted with several por-

tions of THF and stored in the dark at -10°.

Preparation of Titanocene Monochloride

Although preparative methods using aluminum or zinc25’26

for reduction of titanocene dichloride give good yields
of the monochloride, they are unsuitable for use with

the beads. Procedures using i-PngBr and l,A-dioxane, or
Et2AlCl were modified slightly to be consistent with the
syringe techniques used with the beads. The green-brown
Cp2TiCl was purified by recrystallization from toluene.
An esr singlet at g = 1.979, width 3.56,gave a charac-
teristic pattern in the frozen solution. Titanocene
monochloride obtained by i—PngBr reduction was often

used directly for further reactions.

Preparation of Benzyl Titanocene

The benzyl derivative was chosen for its thermal

stability compared to the phenyl derivative. The procedure

18

of Teuben and Meijer27 was followed. Brown crystalline
product was obtained that had an esr singlet at g = 1.983,
width lAG in ether. A frozen solution esr spectrum had
a signal at g = 1.983 and a pattern similar to titano-

cene monochloride.

Nitrogen Fixation Using Benzyl Titanocene

28

The procedure used was a model for attempted nitro-
gen fixation by using the polymer supported species. A
sample of 0.38 gm Cp2TiC1 (1.8 mmol) was placed in a flask
with a septum side arm and 50 mL of dry ether added.
Benzyl titanocene was prepared under nitrogen at -20°C

by slow addition of A.5 ml of PhCH MgBr in ether and 1.0

2
ml of l,A-dioxane by use of a syringe. Reddish brown
insoluble material formed, which, when cooled to -80°C
under nitrogen, developed an intense blue color. Cold
toluene was added, followed by 10 mL of 0.73 M NaNp in THF.
The mixture was stirred and allowed to warm slowly over

a 12 hour period. Hydrolysis of the mixture with methanol
and aqueous hydrochloric acid was followed by separation
of the aqueous layer and a KJeldahl distillation into a

A% boric acid solution. Titration with standard HCl

indicated a yield of ammonia of 0.21 mmol/mmol Ti.

19

Preparation of Polymer Supported Titanocene Monochloride

Polymer attached titanocene monochloride was prepared
by two methods29’3O that were adapted to use with the

bead polymer and syringe techniques.

Method A: A sample of 0.12 gm of polymer attached

 

titanocene dichloride (0.03 mmol Ti) was placed in THF in
a flask. At -70°C 0.03 mmol of i-PngBr was added drop—
wise. It was warmed to -20°C and stirred to obtain green

beads.

esr g = 1.980 57G A peak pattern

far-ir - 153, 191, 386, A38, A82 cm"1

Method B: l g of polymer attached titanocene di-

 

chloride beads (0.2A mmol T1) was placed in a 100 ml
flask and stirred with 10 mL toluene under nitrogen.
When A ml of 25% Et2A1Cl in hexane was added, the beads
turned a brown. The solution was heated to boiling for
one hour and stirred at room temperature overnight to
obtain dark green beads. Excess reagent was removed by
use of syringe and the beads were washed with 2x 10 mL
toluene and 2x 10 mL of ether. Light green beads were

obtained.

esr g = 1.980 A peak pattern

20

Preparation of Polymer Attached Benzyl Titanocene

A sample of 0.53 g polymer attached titanocene mono-
chloride (0.13 mmol Ti) was loaded in a glove box into a
flask with septum side arm. After attaching the flask to
an argon line, 10 mL of THF and 1.5 mL of 0.10 M PhCHZMgBr
in ether plus 0.11 mL of l,A-dioxane was added. The solu-
tion was cooled to -60°C and stirred overnight before
being used to wash the beads with 2 x 10 mL hexane. An
esr of the brown beads indicated a change in the g
(g = 1.983) value and the splitting between peaks from that

of the titanocene monochloride beads.

Treatment of Polymer Attached Benzyl Titanocene with Nitro-

gen

The brown beads from the preparation above were placed
in freshly distilled toluene in an esr flask containing
one atmosphere of nitrogen. After being cooled to -80°C
in a dry ice-acetone bath and being stirred under nitro-
gen for several hours, no color change in the beads oc-
curred. The esr spectrum remained the same as for the
benzyl titanocene. It was concluded that no nitrogen
uptake occurred. No ammonia was obtained after the ma-
terial was reduced with NaNp and hydrolyzed by methanol

and hydrochloric acid.

21

Preparation of Titanocene Monohydride Dimer

Recrystallized titanocene dichloride was placed in a
100 mL flask and the system evacuated to remove residual
moisture. After repeated purges with hydrogen gas, 10
ml of dry, degassed hexane was added to the flask by use
of syringe. The insoluble titanocene dichloride was
suspended in solution by vigorous stirring and 2 equivalents
of n-BuLi in hexane were added slowly by syringe. A
purple compound formed if the reaction was carried out
under hydrogen gas at one atmosphere and at 20° to 0°.
Temperatures above 0° cause a gray insoluble polymeric
hydride to form. In most cases the purple hydride was
used directly for reaction or catalysis because of the
hydride's sensitivity to air and water. The dry solid
could be isolated by careful filtration at temperatures
below -20°. Decomposition to give the green "titanocene"
and a yellow wolid occurred readily. The purple hydride
dissolved with reaction in most other solvents than hexane.
An ir peak at 1125 bd cm'1 disappears when the purple
solid is exposed to air.

The method above could be modified to yield the
purple hydride when other alkyl lithium or Grignard re-
agents were used. Because of the product's instability
in ether solvents, the alkylating reagents must be in
hexane or the ether removed by evacuation from the un-

stirred reaction mixture at -A0°.

22

Reactions of the Titanocene Monohydride Dimer

 

1. Reaction with Nitregen

The purple hydride in hexane was cooled to -80°

and the hydrogen gas removed by evacuation. Dry, de-
gassed toluene at -80° was added and the purple slurry
slowly turned an intense blue. The previously weak esr
signal at g = 1.99 disappeared and the solution became
diamagnetic. Isolation of the blue compound with nitro-
gen was not possible. Upon warming to room temperature
the black color returns. Recooling the slurry to -80°
gave the blue compound again. Under an atmosphere of
argon the purple turned black in toluene but no blue

compound was formed at low temperature.

2. Nitrogen Fixation

The blue complex obtained from the purple hydride
cooled under nitrogen was treated with a ten-fold excess
of NaNp in THF. The temperature was .maintained at -80°
for five hours and then allowed to rise to room temperature.
The mixture was hydrolyzed by a solution of methanol and
0.5 A HCl. Organic and aqueous layers were separated by
using a separatory funnel and the organic layer was wash-
ed with 3 x 25 ml acid solution. Ammonia yield was
determined by Kjeldahl distillation into A% boric acid

solution and titration with 0.1 A standard HCl.

23

3. Reaction with Deuterium Chloride

The purple hydride, (Cp2TiH)2, was prepared under
hydrogen by using BuLi to reduce titanocene dichloride.
A flask containing the slurry of the purple material in
hexane at -A0° was evacuated and repeatedly purged with
argon. After the solution was frozen at -196° the flask
was evacuated and a measured quantity of deuterium chloride
admitted. The slurry was warmed to 0° and stirred. The
purple slurry turned red as the insoluble titanocene di-
chloride formed. After the reaction was complete, the
flask was immersed in liquid nitrogen to condense all
volatile vapors and a sample of the hydrogen gases was
collected in a tube designed to fit the mass spectrometer.
Analysis of the hydrogen isotopes was achieved by compari-
son of the peaks in the mass spectrum or "spectra" if

more than one.

A. Isomerization of Dienes

Allylbenzene and 1,5-cyclooctadiene were dried with
barium oxide, distilled at reduced pressures, and stored
under argon. Products from the isomerization reactions
could be analyzed by integration of the nmr spectrum or
by gas chromatography separation. A ten foot column of
10% SE-30 on Chromosorb P was used in a Varian model 920

gas chromatograph.

2A

5. Isomerization of leS-cyclooctadiene

A sample of 0.1A mmol of the purple hydride, (CpZTiH)2,
in 10 mL of hexane was evacuated, flushed with hydrogen,
and placed under an argon atmosphere. To the purple
slurry at 0° was added two ml of 1,5-COD by use of a
syringe. The mixture was stirred at 0° for 2A hours with
periodic samples taken for analysis. Isomerization to

1,3-COD is essentially complete after seven hours.

6. Isomerization of Allylbenzene

A sample of 0.072 mmol of purple hydride in 10 mL
hexane was evacuated and flushed with argon to remove
hydrogen. To the purple slurry at 0° was added two mL
allylbenzene and the mixture stirred for 2A hours with
periodic sampling. Isomerization to cis-propenylbenzene

and trans-propenylbenzene was complete within 13 hours.

7. Isomerization of 1-Octene

A sample of 0.09 mmol of purple hydride in ten mL
hexane was evacuated and flushed with argon. Two mL of
l-octene was added and the mixture stirred at 0° over-
night. After 12 hours the slurry was yellow and 5%

isomerization to 2-octene had taken place.

25

Reaction with HCl

Hexane solutions of the purple hydride reacted with
anhydrous HCl to form an insoluble red material that was
identified as titanocene dichloride. Approximately 80%
of the original amount of Cp2T1012 could be recovered by
filtration of the red solid from the hexane solution

followed by recrystallization from toluene.

Reaction with Carbon Monoxide

 

Hexane solutions of the purple hydride reacted at one
atmosphere of CO to form a red—brown solid. The brown
solid had ir peaks at 1985 and 1905 cm'l. Under these
mild conditions titanocene dicarbonyl, Cp2Ti(CO)2, was

formed.

Reaction with Triphenylphosphine

 

A hexane solution of (Cp2TiH)2 was allowed to react
With a 3-fold excess of PPh3 dissolved in hexane at -80°.
A translucent blue solution formed immediately. An esr
spectrum at -80° indicates formation of Cp2Ti(PPh3)H
(Figure2l). When the temperature of the solution was in-
creased, the blue faded to brown and an esr signal at

g = 1.97 appeared.

26

Preparation of Titanocene Monohydride From the Dimethyl

Derivative

Method A: Dimethyl titanocene was prepared by the

method of Class and Bestian214 as a yellow solid. When

 

placed in hexane under one atmosphere of hydrogen gas

at room temperature, the purple hydride was obtained after
an induction period of several hours. The purple compound
became gray-black after a few minutes, if the reaction

temperature Was maintained at 25°.

Method B: The purple hydride can be obtained more

 

easily by combining the dimethyl preparation and the treat-
ment with hydrogen. Titanocene dichloride was suspended

in hexane at 0° under one atmosphere of hydrogen gas and

2 equivalents added of methyl lithium in hexane. A yellow
compound formed immediately which yielded the purple hydride
after an induction period of 1-3 hours. Catalysis by using
the purple compound prepared by this method gave the same

results as when BuLi was used as the reducing agent.

Hydrogenation by Using Titanocene Monohydride Dimer

The monohydride catalyst was prepared in the hydro-
genator and used directly. A 0.0121 g sample (0.0A9 mmol
T1) was placed in a 100 m1 round bottom flask with a

stopcock side arm fitted with a septum. The reaction

27

flask was fitted to the hydrogenation apparatus evacuated,
and flushed with hydrogen to remove air and water. To the
flask was then added 20 mL of dry hexane and the solution
was cooled to 0°. Slow addition of 0.1 mmol of BuLi in
hexane by use of syringe initiates reaction. The red,
insoluble titanocene dichloride darkened and a slightly
soluble purple complex formed within an hour. This slurry
of the purple monohydride is used directly for the hydrog-
enation by merely adding the desired olefin syringe.
Hydrogenation began immediately when 1.0 mL of 1-
hexene is added to the slurry of catalyst held at 0°.
Gas uptake measured by using a mercury filled buret to
balance the hydrogen pressure to atmospheric pressure.
The reaction was complete within 5 minutes with an over—
all hydrogen uptake of 1 mmol hydrogen per mmol 1-hexane.
Hydrogenation rates are taken as the maximum slope on the
pseudo first order rate curve. A rate of 923 mL/min-mmol
T1 was observed for the first addition of l-hexene. Sub-
sequent olefin additions led to progressively decreasing

rates of hydrogenation.

Hydrogenation using Titanocene Dihydride Anion

A sample of 0.2A7 g titanocene monochloride (0.12
mmol Ti) was suspended in 20 m1 of degassed THF under
argon. By use of syringe A.2 mL of 1.1 M i-PngBr in THF

was added to the slurry at -80°. After being stirred for

28

two hours the black slurry was warmed to 0° and used for
hydrogenation. A 1.0 mL sample of l-hexene was added

and the gas uptake observed. A rate of 8.5 mL/min-mmol
Ti was obtained. An esr spectrum of the translucent black
solution was a 1:2:1 triplet (g = 1.993, width - 7.2 G)

which is characteristic of the (Cp2TiH2)-.13

Hydrogenation with BuLi Reduced Homogeneous Titanocene

Dichloride

A 0.0127 g (0.051 mmol) sample of Cp2TiC12 was placed
in a 100 mL flask attached to the hydrogenator. After
the flask was evacuated and flushed with hydrogen, 10
mL of dry hexane and 1.0 mL of 1.6 M BuLi in hexane were
added at room temperature (23°). A gray-black slurry
develops within 30 minutes. Hydrogenation occurred when
0.5 mL of cyclohexene was added to the test solution at
25°. A hydrogen uptake rate of 0.12 mL/min was observed,
which corresponds to an overall rate of 2.A mL/min-mmol
Ti. A similar experiment with l-hexene yielded a rate of

160 mL/min-mmol Ti.

Reduction of ®-Cp2TiCl2 and Treatment with DCl

Polymer supported titanocene dichloride beads (2.0 g)
were weighed out in the reaction flask. After evacuation

and purging with argon, the flask was charged with freshly

29

distilled hexane (10 mL) and BuLi (2mL of 2.A M in hexane).
The beads changed from red to black within 10 minutes at
room temperature. The mixture was stirred overnight to
ensure complete reaction, then the beads were washed with
hexane (2 x 10 mL) and with THF (3 x 10 mL). THF (A mL)
was added and the contents of the flask frozen in a liquid
nitrogen bath. The sample tube was attached and the
system evacuated. DCl was added quickly to the frozen
system to ca. 1 atm and the reaction flask closed off.

The contents were then warmed to the desired reaction tem-
perature. Bubbling could be observed and a slow change in
color from black to pink beads occurred. After an interval
of 1 to 1A hours the contents were once again frozen and
the sample tube closed off. Analysis of the gases within
the tube was made by use of a mass spectrometer. Values

are given in Table in the Results section.

High Pressure Nitrogen Fixation

The Volpin and Shur method31 for nitrogen fixation,
which uses a mixture of Cp2TiC12 and PhLi in ether under
100-200 atm of pressure was followed. All reactions were
carried out using 1200-1500 psi of nitrogen in an auto-
clave at ambient temperatures and with continual stirring
during the reaction period. After the Cp2TiC12 or @-

CpTiCpCl2 was loaded into the autoclave, it was flushed

30

with nitrogen and then twice pressurized to 1000 psi and

the pressure released before the reaction was begun.

Nitrogen Fixation by Use of Homogeneous CpQTiC1Q

A 0.651 g sample of recrystallized Cp2TiCI2 (2.61
mmol) was loaded into a glass liner and placed into the
autoclave. After purging the autoclave of air, 100 mL of
diethyl ether and 8 mL of 1.7 M PhLi in ether were intro-
duced using syringes. A slight positive pressure of nitro-
gen was maintained during the loading procedure. The mix-
ture was stirred at ca. one atmosphere pressure for one
hour and then the autoclave pressurized to 1500 psi.

After 15 hours the system was depressurized and the mix-
ture acidified with 20 mL of CHBOH and 20 mL of 20%

H280“. Following a reaction period of 30 minutes, the
glass liner was removed and the volume of solution re-
duced by evaporation. The solution was made basic and

was ammonia steam distilled into a A% boric acid solution.
Titration with 0.1032 M HCl by use of bromocresol green
as an indicator indicated the amount of ammonia. A total
of 1.80 mmol NH3 was found for a yield of 0.69 mmol/mmol
Ti.

31

Nitrogen Fixation by Use of Polymer Supported Cp2T1C1pEE

 

The reaction was repeated by using 3 g (1.8 mmol Ti)
of polymer attached Cp2TiCl2. The beads were loaded in
the glass liner, the system purged with nitrogen, and 5 mL
of 1.8 M PhLi in 50 mL of ether introduced using a syringe.
After one hour of stirring at atmospheric pressure, the
system was pressurized to 1200 psi for 16 hours, then
depressurized and acidified with 20 mL of CH3OH and 20 mL
of 20% H2SOu. Powdered beads were removed by filtration,
the solution volume reduced by evaporation, and a Kjeldahl
distillation into A% boric acid performed. Titration with
standard HCl indicated that no ammonia had been formed.
Table in the Result section summarizes the results of

several attempts at nitrogen fixation.

RESULTS AND DISCUSSION

Polymer supported titanocene dichloride originally
prepared by C. Gibbons,19 has been shown to be an active
catalyst for hydrogenation,15 isomerization, and other

1“ Before use the supported titano-

catalytic reactions.
cene dichloride must be reduced to form the gray or gray-
green beads which contain the active catalyst. The exact
nature of the active catalyst is still unknown. When
alkyl lithium, sodium napthalide, or Grignard reagents

are used, an air sensitive, paramagnetic species is formed
on the polymer. The reaction conditions and the g value
of the esr signal are consistent3 with the active species
being a Ti(II) or Ti(III) compound. Although several

possible compounds could be formed, the reactive compound

is predicted to be the titanocene analog, I,

 

 

32

33

Like titanocene, the analog would be expected to exhibit
3A.35

carbene like behavior toward nucleophilic molecules.
Titanocene is reported to undergo a large variety of
reactions.36 The reaction of nitrogen gas with titano-
cene or titanocene derivatives is especially interesting.
Only a few metal compounds37 react with the normally inert
nitrogen molecule, but in several instances titanium
complexes have reacted with nitrogen to form ammonia.38
Generally the titanium-nitrogen-fixation systems have

not, with a few exceptions,39’140

been catalytic, however.
At first, several forms of "titanocene" were reported,

not all of which reacted with nitrogen. The inactive

form has now been identified17 as a dimeric molecule with
a fulvalenide bridge between the titanium centers. The
active form is generally presumed to be Cp2Ti or the dimer

37,Al,A2

(Cp2T1)2. A number of researchers favor the dimer

form as the species that actually binds the dinitrogen.
A3

Bercaw recently published a crystal structure for
{[C5(CH3)S]2T1N2}2N2 and indicates that a dimeric titano-

cene is probably the species which binds with nitrogen.

Ti: ‘Ti

MW

3A

Formation of the fulvalenide bridged "titanocene" is
a deactivation mode for homogeneous (hS-CSH5)2Ti that is
not possible with the polymer supported species due to
separation of the metal sites by the polymer. Titano-
cene is predicted to be an unstable species with a carbene-
like reactivityuu’35 unlike the reactivity of the other 3d
metallocenes. Singlet carbenes are known)45 to undergo
facile insertion into C-H 0 bonds, addition to n bonds
of alkenes and alkynes, and addition to n acceptors like

A6

C0 and even N2. Besides accounting for reactions of
titanocene with other compounds, the titanocene carbene-
like behavior leads to rearrangements within the mole-

cule itself. Thus a o-hydrogen abstraction from a cyclo-

pentadienyl ring may occur to form a (hl-CSHu)(h5-C5H5)T1H

 

 

structure I1136 Intermediate III then could dimerize with

35

a second titanocene molecule to form the inactive ful-
valenide bridged dimer, II. Even the more stable bis(penta-
methylcyclopentadienyl) titanium (II) undergoes a related
reaction by abstraction of a hydrogen from a ring methyl

group.

<3 = 5‘7" .. 2":
{a 5% ‘5 mgr.

IV .v

Because of its carbene-like reactivity, titanocene
is best considered an intermediate. Besides the hydrogen
abstraction reaction, which leads to an inactive dimer,
another dimerization process is possible to form
[(hs-CSH5)2T1]2.36 The presence of normal hS-CSHS
rings is supported by its chemical behavior. When ex-
posed to HCl gas, solutions of (Cp2Ti)2 rapidly form
CPZT1C12 or if (Cp2Ti)2 is exposed to CO, the dicarbonyl
complex, TiZTi(CO)2, is formed. By contrast, the green
[(95H5)T1H]2(010H8) reacted with HCl to form the magenta
fulvalenide bridged monochloride dimer, [(05H5)TiCl]2-
(C.L()H8).M’u8 Also, unlike the inactive dimer, (CpZTi)2

is in equilibrium with the titanocene monomer as indicated

36

lllcs I

wa/ v

by paramagnetism of toluene solutions of the dimer. Two
especially significant reactions of (Cp2Ti)2 solutions

are the reaction with D2 and the reaction with nitrogen.
Under D2 gas a complete exchange of deuterium between the
gas phase and all positions on the cyclopentadienyl rings
occurs. This exchange implies a ring hydrogen abstraction
process to form a hydride intermediate. Under N2 the

dimer reacts to form (Cp2Ti)2N2. This compound can produce
ammonia and suggests that a dimer is required for the
nitrogen fixation reaction.

N
5 2
2(h -C5H5)2Ti-——+(Cp2Ti)2-————*(Cp2Ti)2N2

By reacting solid dimethyl titanocene, Cp2Ti(CH3)2,
A9

I

with hydrogen Bercaw succeeded in isolating a violet

compound which he identified as the titanocene monohydride

37

dimer, [(hS-C5H5)2TiH]2. A diborane-like double hydrogen
bridge between the two metal centers has been postulated.
This postulate is supported by infrared and mass spec—
trometry. The reactions of the monohydride dimer are

very similar to those of titanocene (Figure 2 and Figure 3).
E H
Ti \

<.

Titanocene monohydride dimer

Ti

6’
WV

Since the monohydride dimer can form the titanocene dimer,
(Cp2T1)2, by loss of hydrogen, titanocene is probably an
intermediate in the reactions of the monohydride dimer.

The hydride bridged dimer may also polymerize to form a
gray-green hydride, (Cp2TiH)x, which is a convenient source
of the reactive form of titanocene. The ir band at 1A50
cm'l, which is associated with the diborane-like hydride
bridge, shifts to llAO cm"1 for the gray-green polymer,
(szTiH)x. It is significant that the hydride dimer and
polymer retain the n bonding to the rings and therefore

undergo the reactions characteristic of (hS-C5H5)2Ti

rather than the inactive fulvalenide bridged structure,

deactivation

 

‘IT3: (Csfl’fifl)zclofla

1,2 “2. o.c s -
"dry" I 1’: [(h -Csfls)211fllz

 

1/2 H2. 0'c

:— 5
It hexane 1" [(h ‘C’fls)ztifll‘

"\m
1/2 N2 uzo

so”: NH, + Ti(IV)

 

 

 

 

' hexane i 1’2 Icio"1o“"c6"s’3'2

Figure 2. Reactions of titanocene.

deactivation

.‘ ‘4‘ (C’Hsflfl)zclofla
‘ CO J_‘_ S
2 (h -C5H5)211(C0)2
a
D

 

 

 

 

 

Q“? - “2 z a 2 ‘ (0:5 c u ) no)
It It It ’ 3 3 2 3
CLE/ <5
L .

n no

2 _....._.2 2 m + 2 nuv)

aolvent 3

205:

 

II=_ 5
2 (h -Csfls)ztl(l)Cfl3

Figure 3. Reactions of titanocene monohydride dimer.

39
[(hS-C5H5)TiH]2(ClOH8).

Hydrogenation Studies

Butyl lithium reduced polymer supported titanocene
dichloride have been tested for catalytic activity towards
hydrogenation of unsaturated organic compounds.13 With
the olefins cyclohexene and l-hexene, the rates of hydro-
genation were several times greater for the polymer sup—
ported titanocene species than a homogeneous titanocene
species prepared under similar reaction conditions. The
active catalyst is the gray copolymer beads that are
paramagnetic and air sensitive. Repeated additions of
olefins can be made without significant reduction in
hydrogenation activity. When hydrogenation is completed,
the polymer attached titanocene dichloride may be re-
generated by treatment with HCl. Therefore, it is concluded
that the bis-w—cyclopentadienyl integrity is retained
throughout the reduction procedure.

The active titanocene species can only be surmised
from such indirect evidence. It has been observed that
1 or 2 electron reductions are possible and that when an
excess of reducing agent is used, no difference in either
esr signal or hydrogenation rates is noted. Reduction of
titanocene dichloride with alkyl lithium reagents would

be expected to follow the sequence below:

A0

R R H
Cp TiCl __.,Cp Ti _.Cp Ti ——>Cp Ti
C Ti/H c T' + H
p2 \H—* p2 l 2
Scheme (1)

Cp2TiC12 JELCp2TiC1—+Cp 2Til-R —->Cp2Ti-H

Scheme (2)

The concentration of reducing agent, solvent, temperature,
and type of inert gas used are all factors that determine
which of the several possible products are obtained.
Hydrocarbon solvents would tend to favor scheme (1)
whereas ether solvents would favor formation of Ti III
compounds, Cp2Ti-R or Cp2Ti(H)2'. Unstable sigma bonded
alkyls with B-hydrogens or the unstable dihydride are
known to decompose with loss of an alkane, alkene, or
hydrogen gas.50 Thus, the catalytically active species
on the polymer might be the titanocene analog, (1),

or might be a hydride (VI) or (VII).

The effect on the hydrogenation rate was determined
when the polymer supported catalyst was prepared by use
of other reducing agents: i-PngBr, NaNp, CH3Li, C2H5MgBr.
Little difference in the hydrogenation rates was observed

when BuLi was used for reduction. This result implies

a

0&1 Quin/H @:n<“ H
d e ‘ 8 ,.

that the same catalytically active species is being formed
on the polymer in each case.

When polymer attached titanocene derivatives, Cp2TiCl,
‘szTi(CH2), are reduced with an excess of reagent, the
hydrogenation rates are also nearly the same as when
reduction is of the polymer attached titanocene dichloride.
Although beads reduced under hydrogen generally have
higher hydrogenation rates, it is not known whether the
problem is due to handling technique or a difference in
the species formed. Experience with the titanocene
hydride derivatives leads one to suspect the former

explanation. In all cases the esr signal of the reduced

A2

species is the same.

Comparison of the Beads with (Cp2TiH2):

 

Reaction conditions known to lead to specific titano-
cene derivatives were used to compare hydrogenation rates
of the polymer catalyst with the homogeneous catalyst. When
a A0-60 fold excess of an alkyl lithium or Grignard reagent
is used in an ether solvent at less than room temperatures,
the Ti (III) alkyl or dihydride anion compounds are formed.
The process can be observed by esr. The dihydride anion,

(Cp2TiH2)', is stable for days at room temperature. For
Cp2TiCl2 + xs i-PngBr _+(Cp2TiR2)‘__s(Cp2TiH2)'

the homogeneous species a hydrogenation rate for l—hexene
was 8.5 mL/min-mmol Ti whereas the polymer attached species
under similar conditions had the same hydrogenation rate
observed before, 38 mL/min-mmol Ti. Since the dihydride
anion is not believed to undergo a deactivating dimeriza-
tion reaction, the higher hydrogenation rate by the bead
catalyst indicates a different species than the dihydride
anion is formed on the polymer.

"Preparative methods for the dihydride anion specify
an ether solvent.50’51 The polar solvent is undoubtedly
involved in solvation and stabilization of the anion.

Another titanocene alkyl anion, (Cp2TiC2H5)2-6MgCI2-7(02H5)2O

A3

prepared by Marvichu7 under similar conditions requires
an ether solvent. Thus, when a A0 fold excess of i—
PngBr is added to Cp2Ti012 in a large excess of hexane,
the mixture turns brown instead of black and the dihydride
anion does not form as indicated by the lack of its trip-
let esr signal. Since the titanocene dichloride beads
are normally reduced in hexane and the esr signal of the
beads does not resemble the frozen glass spectrum52 of
(Cp2TiH2)', one would not expect the polymer supported
species to be the dihydride anion.

Another possibility is that the supported titanocene
species is the monohydride reported by Bercaw.‘49 The
monohydride is normally a hydride bridged dimer, but a

weak esr signal and the compound's reactivity indicate a

monomer-dimer equilibrium. Preparation of the mono-
hydride dimer is effected by the reaction of the solid
dimethyl titanocene-with hydrogen gas.u9 This same re-

action, if carried out in a hydrocarbon solvent, leads

AA

Cp2T1(CH3)2 + 3/2 H2-———»1/2(Cp2TiH)2 + 20H,

to green "titanocene", (CpTiH)2(ClOH8). Although not
isolated, the dimer (Cp2TiH)2 was probably formed as an

intermediate in the reactions reported by two authors.u7’53

Preparation of Titanocene Monohydride Dimer

A convenient method for the preparation of the mono-
hydride dimer, (Cp2TiH)2, in high yields in a hydrocarbon
solvent has now been developed. Titanocene dichloride
may be reduced in hexane under one atmosphere of hydrogen
to the purple monohydride dimer by use of two equivalents
of BuLi. Care must be taken to maintain the reaction
temperature between —20° and 0°. The insoluble red
titanocene dichloride is observed to darken slowly and a

hexane

Cp2Ti012 + 2BuLi + H -————-+(Cp2TiH)2 +

2

2CuH10 + 2L1C1

slurry of slightly soluble purple (Cp2TiH)2 to form, as
the reaction mixture is stirred. Reaction is complete in
one-half to one hour of stirring of the mixture at 0°.

The purple solid may be isolated by filtration with
Schlenkware followed by washing the solid with cold hexane

or pentane. Non-hydrocarbon solvents or temperatures

AS

above 0° cause the purple hydride to turn green and even-
tually yellow. The preparative reaction may follow one of
the reaction schemes below.

H
2
Cp2TiC12 + 2BuLi-—>~Cp2Ti(Bu)2-———-..1/2(Cp2TiH)2 +

2CuH10

Scheme (3)

BuLi H2
Cp2TiCI2 + BuLi—pCp2T101 —_—>CpTi-Bu -———>

1/2(Cp2TiH)2 + cquO

Scheme (A)

Reaction scheme (3) follows that of the dimethyl titano—
cene, except that the butyl groups have B-hydrogens that
may transfer to the Ti by B-hydrogen elimination. Only
titanocene alkyl groups without B-hydrogens are reasonably
stable at room temperature. Thus the less stable dibutyl
derivative reacts at low temperature in much the same way
as the dimethyl derivative and forms the same product,
(CpZTiH)2.

Reaction scheme (A) is also possible. Titanocene
monochloride can be prepared by slow addition of an

equivalent of i-PngBr or BuLi to the dichloride. Mono

A6

alkyl compounds can be prepared from the monochloride,
although only alkyl groups without B-hydrogens are stable
at room temperature.5u The possibility of this reaction
sequence was demonstrated in the lab by reaction of solid
Cp2TiC1 with one equivalent of BuLi in hexane under a
hydrogen atmosphere. The purple monohydride dimer is
formed if the temperature is 0° or below.

Other reducing agents may be used to prepare the purple
hydride if the agents have B-hydrogens. Thus i—PngBr
and C2H5MgBr can also be used to make the purple dimer.
Because the dimer reacts with most solvents other than
hydrocarbons, the Grignard reagents must be placed in
hexane before addition to the titanocene dichloride. The
monohydride dimer did not form when NaNp was used. Other
reducing agents that operate by alkylation followed by
elimination should also be useful for this reaction.

When dimethyl titanocene is stirred in hexane under
hydrogen the slurry turns purple, but within several
minutes the green fulvalenide bridged dimer forms.2u
Bercaw, therefore, isolated (Cp2TiH)2 by reaction of the

36 It was found that

solid Cp2T1(CH3)2 with hydrogen.
titanocene dichloride will react with 2 equivalents of

CH3Li under hydrogen to give the purple hydride in hexane.
This procedure simplifies preparation because the dimethyl
intermediate need not be isolated. Upon addition of CH3Li

the red Cp2TiC12 turns to a yellow slurry of Cp2Ti(CH3)2.

A7

After 2-5 hours of stirring, the yellow slurry will turn
completely purple within 2-5 minutes of the first faint
sign of purple. The purple slurry of (Cp2TiH)2 is then
as stable as the dimer when prepared from BuLi. If the
yellow Cp2Ti(CH3)2 slurry is stirred in the dark under
hydrogen, the purple compound is not obtained. A photo-
chemical reaction mechanism is implied by this observa-
tion and is consistent with literature reports.5u’55
The monomeric titanocene monohydride is also a pos—
sibility for the structure of the reduced polymer sup—
ported catalyst (Structure VI). Originally, (Cp2TiH)2
was prepared by reaction of Cp2Ti(CH3)2 with hydrogen,“9
but can be prepared more easily by BuLi reduction of
Cp2TiCl2 under hydrogen. The purple hydride, (Cp2TiH)2,
was tested as a hydrogenation catalyst so that hydrogen

uptake rates could be compared with those of the reduced

polymer supported catalyst.

Hydrogenation by (CpDTiH)2

Hydrogenation rates obtained with the purple hydride
and l-hexene are approximately 30 times faster than the
reduced beads and hundreds of times faster than the
homogeneous Cp2TiCl2 reduced at room temperature, the
species used for previous comparisons with the bead
catalyst.

The purple hydride, (Cp2TiH)2 was prepared in a flask

A8

attached to the hydrogenation apparatus and olefin added
by syringe to initiate hydrogenation. Results of testing
several olefins for hydrogenation rate are given in
Tables 1 and 2. Generally 0° was found to give the high-
est reproducible rates without the gray insoluble polymer
(CpZTiH)x being formed. Selectivity for primary alkenes
over those with internal was large bonds. Table (3)
compares hydrogenation rates of olefins at two different
concentrations. Doubling the total amount of olefin

at the same concentration gives comparable results, but
more reproduciblerates‘were obtained when 1.0 mL of
olefin was added to 20 mL of solvent and catalyst. Rates
of hydrogenation obtained with (Cp2T1H)2 were greatly in
excess of any obtained with either the bead system or any
homogeneous titanocene systems tested before in these
labs. Even with vigorous stirring the systems were
hydrogen diffusion limited. The rates are affected by
stirring rate and by Ti concentration (Table (A)). The
rates in Table (1) are, therefore, not maximum rates,

but rather rates obtained under conditions comparable to

those used when the reduced titanocene beads were used.

Effect of Temperature on Hydrogenation Rates

Figure (A) shows a plot of reaction temperature mg.

rate of hydrogenation of l-hexene and cyclohexene with

Table 1. Hydrogenation Rates Compared:

A9

Species and (Cp2TiH)2.

Polymer Attached

 

 

Hydrogenation RatesC

Polymer Attacheda

 

 

 

Olefin Titanocene Species (Cp2TiH);)
l-hexene 213.0 970
1-octene --- 950
2-octene --- 170
Styrene 2A3.0 70
cyclohexene 90.3 20
l-methylcyclohexene 1.0 ---
1,2-dimethylcyclohexene 0.0 ---
1,3-cyclooctadiene 216.0 ---
1,5-cyclooctadiene 183.0 705
isoprene --- 60
l—hexyne (polymer) ---
3-hexyne 1A9.0 ---
diphenylacetylene A0.6 210
aAt 25°.
bAt 0°.

C(mL H2/min-mmol Ti).

50

Table 2. Hydrogenation Rates13 at 25 °C for Cyclohexene.

 

 

 

Rates
(mL H2/min-mmol)

Catalyst Precursor mmol-Ti Initial Maximum
titanocene dichloride 0.05 5.6 28
attached Cp2TiCl2 0.03 88.7
Attached Cp2TiC12
ground before reduction 0.03A 105.6
Benzyltitanocene dichloride 0.2 9.5

titanocene dichloride 0.2 10

 

 

51

Table 3. Effect of Olefin Concentration on Hydrogenation

 

 

 

Rates.
Rate
mmol Ti Temp. Olefina (mL/min-mmol)
0.0A9 0° 1-hexene 920
0.057 0° cyclohexene 10
0.0A7 0° l-octene 970
0.0A7 0° 2-octene 20
Olefinb

0.056 0° 1-hexene 970
0.0A9 0° 1-octene 770
0.133 0° cyclohexene 20

 

 

a1.0 mL olefin in 20 mL hexane solvent.

b0.5 mL olefin in 10 mL hexane solvent.

52

Table A. Effect of Ti Concentration on Hydrogenation

 

 

 

Rates.
a Rate
mmol Ti Temp (mL/mmol-min)
0.1020 0°C 600
0.0313 0° 775
0.0129 0° 780
0.0108 0° 150

 

 

aIn 20 mL hexane.

53

near-

IOOO--—

900—

soo—

700‘

600-

lm Hzlmln-mmol Til

500—

 

RATE

400-7

 

 

300--
200——
I”
z
/
cyclohexene /”
loo——
\.
\.

l l i L J l l \J; l

l l I I 1 I T I.

-30 -20 -IO .0’ IO 20 3O 40

TEMPERATURE

Figure A. Graph of hydrogenation temperature versus rate.

5A

the purple hydride catalyst. Projection to room tempera—
ture gives rates that might be compared with those of the
bead catalyst. For the purple hydride, the rate would

be in excess of 1,100 mL/min—mmol Ti compared with a
hydrogenation rate on the same apparatus for the titano-
cene beads of only 37 mL/min-mmol. This difference amounts
to a 30 fold rate increase for primary olefins. For
cyclohexene even the projected room temperature rate for
the (Cp2TiH)2 catalyst is lower than the rate with the

bead catalyst. The effective temperature range for hydrog-
enation by the reduced titanocene beads has not been
determined at temperatures above 20° because of polymer-
ization to (Cp2TiH)X or deactivation by formation of a
yellow solid from which Cp2TiCl2 cannot be regenerated

by HCl treatment.

Effect of Incremental Olefin Addition

With each incremental addition of olefin the hydro—
genation rate decreases when the purple hydride is used
as the catalyst. After each 1 mL addition of olefin has
been hydrogenated another 1 mL addition is made. The
olefin concentration decreases only A% per addition whereas
the hydrogenation rates decreased by 10—A0% per increment.
Deactivation of the catalyst rather than dilution is the

cause of the rate decrease. The rate decrease probably

55

Table 5. Effect of Incremental Olefin Additions on
Hydrogenation by (Cp2TiH)2.

 

 

 

 

# of Olefin Rate
Temp. Olefin Additions Conc. (mL/min-mmol)

0°C l-hexene 1 0.381 N 920
2 0.36A 8A5

3 0.3A8 690

0°C l-octene 1 0.305 970
2 0.291 670

3 0.278 A00

A 0.267 260

 

 

56

is due to impurities from each olefin addition that react

with the monohydride dimer.

Effect of Adding PPh3 or THF

Either solvents of triphenylphosphine will react with
(Cp2TiH)2 to cleave the dimeric hydride and give monomic
hydrides. The reaction is examined in more detail in the
esr section. The monomic hydrides plus coordinated THF
or PPh3 were tested for catalytic activity. The results
of hydrogenation of l-hexene are shown below (Table 6).
The species Cp2Ti(PPh3)H is produced by adding PPh3
in cold hexane directly to the cold solution of purple
hydride. A translucent purple solution formed immediately.
Catalytic activity of the purple hydride solution at -20°
was greatly diminished by both PPh3 and THF. The beads
show a similar, though less drastic, effect when treated

with THF or PPh3.

Comparison of Hydrogenation Rates with the Bead Species

Since excess reducing agents can lead to different
species than can 2 equivalents, the beads were tested for
hydrogenation ability after reduction by 2 equivalents of
BuLi. Unreacted cyclopentadiene groups on the polymer
may react with BuLi, so 2 equivalents BuLi/Ti and 1
equivalent per unreacted ® -CpH was added. Conditions

similar to those that lead to the homogeneous purple

57

Table 6. Effect of THF or PPh3 on Hydrogenation Rates.

 

 

 

mmol Ti Temp. Olefin Rate (ml/min-mmol)
0.065a -20°C l-hexene 600
Added PPhB/hexane " 80
Added THF " 20
0.07Aa —20°c l-hexene 620
Added PPh3 " 80
Added THE (brown) " 0
0.056a 0°C 1-hexene 360
Added THF 0°C " 185
0.0A5 (beads)b 0°C l-hexene A0
Added PPh3 0°C " 20
Added THF 0°C " 25

 

 

a(Cp2TiH)2 in hexane.

bBuLi reduced ®-Cp2T1012.

58

hydride were used. As Table (7) shows, 2 equivalents of
reduction are enough to give the beads full activity.

The rate is the same as was obtained with beads reduced

Table 7. Hydrogenation with Beadsa Treated with 2 Equiva-
lents of BuLi.

 

 

 

Prep. Conditions Hydrogenation
Equivalent of b
Temp. Gas Reduction Olefin Temp. Rate
0°C H2 2 eq. BuLi/Ti l-hexene 0°C 37
0° H2 2 eq. BuLi/Ti+()—CpH l-hexene 0° 30
0° H2 2 eq. BuLi/Ti+()-CpH l-hexene 0° 36

 

 

a‘Q’D-Cp2TiCd2 copolymer beads.

b(mL/min-mmol Ti)

with an excess of BuLi, but much less than the homogeneous
(Cp2TiH)2. The bead color was gray-green, Just as was
obtained previously.

When equal amounts of homogeneous Cp2T1012 and polymer
supported ®-Cp2TiC12 are reduced by 2 equivalents of
BuLi in hexane, combined hydrogenation rates are far
below what would be expected if all the Ti were the
catalyst (Cp2TiH)2. When the beads are washed with hexane
several times, the rate decreases to a value on the order

of the expected rate for the polymer supported Ti only.

59

Table 8. Hydrogenation with Beadsa Plus Homogeneous
Cp2TiCl2.

 

 

Rateb
Temp. (mL/min-mmol)

 

Homogeneous Cp TiCl2 0.0281 mmol

2
plus ®-Cp2TiCI2 0.0277 mmo1 0°C 112 lst run

beads only 0.028 0° 25 2nd run
(washed with hexane)

 

 

a (PD -Cp2TiCl2 copolymer beads .

b1.0 mL 1—hexene.

There are two reasonableexplainations of this experiment:
(1) polymer supported titanocene dichloride and homo-
geneous Cp2TiC12 yield different species when reduced by
BuLi, or (2) the hydride monomer cannot dimerize in the

polymer supported Ti.

Comparison of Catalyst Rates

 

A comparison of hydrogenation rates of l-hexene by
(Cp2TiH)2 and by reduced titanocene beads demonstrates a
significant difference in reactivity. Tables 1 and 2
list rates of hydrogenation obtained by other researchers
in these labs. A number of variables make direct compari—

son of rates difficult. Bead size, Ti concentration on

60

the polymer, rate of stirring during reaction, temperature,
total amount of catalyst used, concentration of olefin,
and purity of reagents used have all been observed to
effect the measured hydrogenation rates. The largest dif-
ference is size of the polymer beads; ground heads have
the higher rates of hydrogenation. In Figure 5 the rates
are compared for (Cp2TiH)2 and for reduced beads. Except
for cyclohexene, (CpZTiH)2 has rates significantly better
than those rates obtained with the bead catalyst. Both
catalysts hydrogenate primary double bonds better than
internal double bonds, but the difference is more dramatic
for the (CpgTiH)2.

Dienes and alkynes are hydrogenated by (Cp2TiH)2 all the
way to alkanes. With cyclooctadiene the internal double
bonds so not seem to diminish the rates as much as observed
in cyclohexene. Diphenyl acetylene has a lower rate than
for the non-conjugated diene. Figure 6 compare hydrogena-
tion rates when the same conditions (except for tempera-
ture) are used for (Cp2TiH)2 and for reduced beads with 1-
hexene. As before, (Cp2TiH)2 has higher rates than the
whole beads. The effect of PPh3 and THF on each of these
catalysts is to decrease the rate but the relative de-
crease is greater for (CpZTiH)2.

Titanium concentration is also seen to effect (Figure
6) the rates for (Cp2TiH)2. Figure 7 illustrates this

effect more clearly. At higher Ti concentration the

61

CH, (cugscmcu, 0 © o-czc—o

........

1000-"'-

voo-L‘

if; 3 [szTi H12

.00 ---a ;.;_L..jf';§f m“ reduced

700--

........

600-—

500 -—

 

.........

lml Hz/mln/mmol Tll

400 ——

RATE

300-—

 

........
.......

...........
... .
..............
“" ... k....
.....................
......

“)0"— éil“???

........

'''''''''''

 

 

 

 

 

 

 

 

 

 

 

 

 

l-hoxono Cyclohoxono 1,5 COD

Figure 5. Relative hydrogenation rates for bead catalyst
and (Cp2TiH)2.

62

700‘"-

600 _ _ 596

500 --

."$-’-?Eif<~‘3‘€:=t-‘$i~fa‘-§§s§
O

n ”in. ‘ .x

400 —-

A'.
~.

lml Hzlmln-rmmol Tll
~¢ffififi$fifififififlfififififi

mm——

RATE

200-

   
 

 

 

Figure 6. Relative hydrogenation rates after THF or
PPh3 addition.

63

.COHumnpcoocoo

 

mflmfleaov mummo> wopmn :oHpmcewopomn mo cameo .5 omswfim
:. 33: o 2
can 00. ca 00 o.v cm at:
0
..oo«
100*
#000
O
1.00»
O
O
#000—
£009

 

BLVU

llJ. IOww-ulw/ZH [my

6A

overall rate tapers off. At low Ti concentration there is
an abrupt drop in rates. This drop is probably because of
limits in purity of reagents and in the ability to exclude
air and water from the system used for hydrogenation.
Because of the Ti concentration effect on observed rates,
most of the (Cp2TiH)2 rates are not maximum rates for the
olefins tested.

The reduced titanocene beads have been compared to the
homogeneous system for hydrogenation of olefins.13 The
bead catalyst was found to be several times more active
than the homogeneous species. This difference is probably
due to decomposition of the homogeneous species; the in-
active dimer, (CpTiH)2010H8, is one of the decomposition
products. Increased catalytic activity was then ob-
served for the bead species that could not deactivate by
dimerization. No attempt was made to maximize the rates
in the homogeneous system. Room temperature is detri-
mental toward formation of active titanocene species,
thus the conditions used assure formation of inactive
species. At lower temperatures a gray material identified
as (Cp2TiH)x forms instead of the inactive brown solution
obtained at 25°. The polymeric hydride, (Cp2TiH)x, is
an active catalyst, although not as active as the purple
dimer, (szTiH)2. The gray polymeric hydride can be
used for hydrogenation at 25°, at which its rate of

hydrogenation is greater than the bead species. Small

65

changes in the reaction conditions and concentrations of
reagents used can change the titanocene species being
formed. It is therefore difficult to make comparisons
of the bead system to a homogeneous system unless the
active intermediate can be identified in each case.

Figure 8 compares the reduced polymer attached titano-
cene dichloride beads to several catalysts that have been
identified in the solutions used for the hydrogenation.
The rates of hydrogenation of l-hexene by the purple
hydride catalyst are much greater than any of the other
catalysts tested. The purple hydride is not as easy to
handle (although easier to prepare) as the reduced beads.

The difference in hydrogenation rates and olefin
selectivity between the purple hydride and the reduced
beads indicates either that the catalysts are different
species or that some severe limitation is being imposed
on the polymer bound species by its polymer attachment.
The most reasonable possibilities for the polymer bound
catalytic species is then polymer attached analogs of
titanocene (I), titanocene monohydride (VI), or fulva-
lenide bridged hydride (II). Reasonable mechanisms can be
drawn for each of the possibilities (Figure 9). If both
the titanocene and the fulvalenide bridged hydride are pres-
ent, it would account for the mixed results when reduced
beads were treated with DCl. In a D2 atmosphere one would

expect deuterium to be incorporated into the ring positions

66

by the titanocene ring hydrogen abstraction mechanism.149
The change in symmetry of the titanocene as hydrogen
abstraction occurs should lead to a second esr signal
superimposed on the titanocene signal. When these species
are bound to the polymer, esr detail would be lost and an
unsymmetrical singlet would be reasonable as the observed

signal.

67

 

 

.H2
m n
O R. 2
a «A P
o by
h
on I
C
v. u 2
c .. OH
H .b
Cd
a...
I
ooM
b.
9
8
Mr
mo.n
..b

 

 

IOOO‘—

 

I...
m
m.
_ — _ _ _ _ _ _
... _ a. ... _ _ . _ a
0 0 O
o o
.... m 7 w m w m ...
.z. .25: Esta: :5 m5;

purple
hYdmlo

+ BuLl

Comparison of hydrogenation rates for titano-

cene catalysts.

Figure 8.

68

 

hydrogenation by titanocene derivat

Figure 9. Mechanisms for

69

Reactions of (QPZTiH)2

 

Catalytic Isomerization of Allylbenzene

The purple hydride was tested for activity as an iso-
merization catalyst. Allylbenzene, ALB, which has a
terminal double bond, can be isomerized to form the more
stable system - SEE? and Emmmee propenylbenzene. The
products were identified by gas chromatography or by
integration of NMR peaks. The purple hydride was prepared
under hydrogen as before, then the hydrogen removed and
argon substituted. Two mL of dry ALB was added to the
hexane slurry and the mixture stirred overnight at 0°.
After 13 hours, 100% isomerization to predominately mmemef
propenylbenzene had occurred. By comparison, the polymer
supported titanocene catalyst had isomerized only 90% of
the ALB after 13 hours at lA5°.lu Unlike the active bead
catalyst, homogeneous (Cp2TiH)2 often turns yellow during

a test and is then inactive.

Isomerization of 1,5-Cyclooctadiene

In a similar manner to allylbenzene, 1,5-cyclooctadiene
(1,5-COD) can be isomerized to the more stable conjugated

1,3-COD. A slurry of (Cp2TiH)2 in 10 mL of hexane was

70

treated with 1.0 mL of 1,5-COD under an argon atmosphere.
At a reaction temperature of 0°, 80% isomerization to
1,3-COD was obtained in 3.5 hours or 100% isomerization in
7 hours. No l,A-COD was obtained, as indicated by gas

1” that the

chromatography. By comparison, Lau reported
titanocene dichloride beads gave 90% isomerization to

1,3—COD after 10 hours at 1A5°.

Isomerization of l-Octene

Unlike the other substrates tested, l-octene does not
have the driving force for rearrangement to a conjugated
diene. The 2-octene is only slightly more stable than 1-
octene. When a slurry of (Cp2TiH)2 was treated with 1.0
mL of l-octene at 0° under argon, 5% isomerization to 2-
octene occurred after 12 hours. Although longer reaction
times might lead to more complete isomerization, the purple
hydride is unstable under argon and decomposes readily to
give an inactive yellow species.

Isomerization reactions of the purple hydride, (CpZTiH)2,
could proceed by the mechanism below if the dimer is in
equilibrium with the coordinatively unsaturated monomer.
The alkene first may bind to the Ti followed by a hydrogen
transfer to an alkene carbon. Two possible B-hydrogen
abstraction reactions could occur. One reaction returns
to the original alkene but the other B-hydrogen abstrac-

tion leads to the isomerization product.

71

Isomeri zat ion QT< Hie/i

Product

When the homogeneous (Cp2TiH)2 catalyst is compared
to the reduced species on the polymer beads, the purple
hydride is a more active isomerization catalyst than the
beads. The stability of the homogeneous compound is
not very stable under argon or at temperatures above 0°.
In most cases the isomerization yield was limited by
decomposition of the purple hydride. By comparison, the'
bead species were active even at 1A5°, which is high for
a titanocene species. Furthermore, the beads could be
reused or regenerated by treatment with HCl. Regeneration
of ®-Cp2TiC12 from the reduced beads, which had been
at 1A5°, indicates the n-cyclopentadienyl integrity had

been maintained. If the polymer supported species were

72
the monohydride monomer, ®CpTi-H it is not clear why the
monomer should have such high thermal stability when
polymer supported.

Nitrogen Fixation

Volpin and Shur High Pressure Method

 

The Volpin and Shur method31 for nitrogen fixation
uses a mixture of titanocene dichloride and phenyl lithium
(PhLi) in ether under 100-200 atm of nitrogen pressure.
Homogeneous and polymer attached Cp2T1C12 were compared
for nitrogen fixation at 150 atm of N2 and ambient tempera-
tures. The ammonia yields for several different experi-
mental runs are listed in Table 9. Homogeneous nitrogen
fixation gave ammonia yields comparable to those reported
by Volpin and Shur.31 When polymer attached Cp2TiCl2
was used, however, no ammonia was detected.

In cases where the titanocene-nitrogen intermediates
have been identified, dimeric titanocene species with
bridging nitrogen molecules were the intermediates that
led to ammonia after hydrolysis.37’uz’u3 The titanocene
sites on the polymer are physically separated and formation
of a dimeric titanocene that could react with nitrogen is
unlikely. Since no nitrogen was fixed by the monomeric
titanocene species on the polymer, the conclusion that a

dimer is required for nitrogen fixation by titanocene is

73

Table 9. Nitrogen Fixation Results: Volpin and Shur

Method.

 

 

 

 

NH3 Yield
meal...
mmol Ti mmol PhLi mmol Ti
Homogeneous Cp2TiCl2 2.A1 13.6 0.73
2.AA 13.6 0.51
2.A3 13.6 0.60
2.61 13.6 0.69
2.A? 13.6 0.67
Polymer Attached
®-Cp2TiC12 1.11 6.5 0.0
1.08 6.5 0.0
1.80 9.0 0.0

 

 

7A
supported.

Fixation of N2 by Cp2Ti(arene) Compounds

A number of Ti (III) compounds of the general formula
Cp2TiR are known to react with nitrogen to form a dimeric
species, (CpQTiR)2N2.56’27’30 Thermal stability of the
Cp2TiR compounds is dependent on the alkyl group. Gen-
erally, substituted aromatic groups give compounds with
the greatest stability.57 Teuben has found that a number
of compounds can react with nitrogen to form blue diamag—
netic complex, (CpZTiR)2N2, which can be reduced by NaNp,
RLi or RMgX to form ammonia and hydrazine.28

The benzyl derivative, Cp2Ti(CH2C6H5), is thermally
stable enough for convenient use in nitrogen fixation.
Preparation results from treatment of Cp2TiC1 with one
equivalent of benzyl magnesium bromide and an equimolar
amount of l,A-dioxane. A brown solid of Cp2Ti(CH206H5)
can be isolated from an ether solution under argon if a
temperature of -20° or below is maintained. In toluene
or hexane the brown compound reacts with nitrogen at -80°
to form [Cp2Ti(CH2C6H5)]2N2, an intensely blue complex.
Formation of the blue nitrogen complex corresponds with
the disappearance of an esr singlet at g = 1.983 and the
mixture becomes diamagnetic. When the blue nitrogen complex
was treated with an excess of NaNp followed by hydrolysis,

0.205 mmol of ammonia per mmol T1 was obtained.

75

The polymer attached compound,(:)-Cp2Ti(CH2C6H5),
was prepared by the above method. Brown beads were ob-
tained that had an esr signal consisting of a four peak
pattern, g = 1.983. The brown beads were treated with

NaNp under one atm N After hydrolysis of the mixture,

2.
no ammonia was detected. When the brown beads were cooled
to -80° in toluene under nitrogen, there was no color
change. The esr signal remained the same. Physical
separation of the Ti sites by the polymer must prevent
formation of the required dimer, (Cp2TiR)2N2, and nitrogen
does not bind to the monomeric compound. This experiment
confirms the conclusion of the Volpin and Shur nitrogen
fixation experiments with the beads. A dimeric titanium
is required for formation of the nitrogen complex and the

polymer support prevents formation of such nitrogen

bridged titanium dimers.

Fixation of N2 by (szTiH)2

 

Bercawug had found that (Cp2TiH)2 could be used to
fix nitrogen. The hydride dimer was treated with NaNp
under 150 atm of nitrogen to obtain a maximum yield of
0.99 mmol NH3 per mmol Ti. It was found in this work
that (CpZTiH)2 fixes nitrogen at one atm N2. The purple
hydride was suspended in toluene under nitrogen,the reac-
tion cooled to -80° to obtain an intense blue solution,

and then a 10 fold excess of NaNp in THF added. After

76

being stirred for 12 hours and being warmed to room
temperature, the mixture was hydrolyzed with CH3OH and
aqueous HCl. The aqueous layer was separated and ammonia
determined by standard HCl titration of the Kjeldahl
distillate. Ammonia yields were the highest when NaNp
was used to reduce the N2 complex. Since the yields were
greater than one, there was the possibility that the
reaction would be catalytic if the reduced nitrogen could
be transferred to a second complex. Lewis acids have
performed this function in one titanium nitrogen fixation

system.“0

The Lewis acid Et2A101 was added to the nitro-
gen complex with NaNp reducing agent, but the yield of
ammonia was lower than before. Other Lewis acids may
function better.

Polymer supported titanocene dichloride was tested
for nitrogen fixation under conditions under which the
monohydride titanocene may have formed. The beads were
treated with 2 equivalents of BuLi under hydrogen in
hexane at 0°. The hydrogen gas was replaced by nitrogen
gas, toluene added, and the mixture cooled to —80°. As
before, the beads remained gray. Sodium naphthalide was
added and, after 12 hours of stirring, the mixture was
hydrolyzed. No ammonia was detected in the aqueous layer.
This experiment confirms that the reduced polymer supported

species does not fix nitrogen. The identity of the reduced

species on the beads in this experiment is not known.

77

Table 10. Nitrogen Fixation Yield by Use of (Cp2TiH)2.

 

 

 

 

Reducing mmol
mmo1 Ti Agent NH3 Yield (mmol Ti)
0.2A NaNp 1.20
0.21 NaNp 1.27
0.25 NaNp 0.9A
0.16 NaNp + Et2A101 ' 0.81
0.38 i-PngBr 0.0
0.37 i-PngBr 0.0

 

0.36 n-BuLi 0.35

 

 

78

DCl Treatment of the Polymer Sppported Titanocene Species

An attempt was made to identify whether the reduced
titanocene species on the polymer was a hydride or not by
analysis of the hydrogen gases liberated when the beads
were treated with DCl. Four runs were made at different
temperatures and the gases analyzed by mass spectrometry.
Table 11 summarizes the reaction conditions and results.

A graph of gas composition mg. temperature shows a straight
line relationship for each of the hydrogen isotopes.

The results of treating the reduced species with DCl
indicates the presence of a hydride, but that not all Ti
centers involve a hydride. There are several ways in
which a hydride could have formed. Bercawug has shown that
ring hydrogen exchange does occur. Brintzinger50 obtained
a Ti (III) dihydride when alkyl lithium reagents were
used where the hydride is from the 8 position on the alkyl
group. The DCl treatment experiment does not answer the
question of the origin of the hydrogen, the oxidation state
of the metal, or the nature of the other ligands.

Presuming that the hydrogen comes from the cyclopenta-
dienyl rings, one can make inferences based on the reaction
temperature ye. percent hydrogen isotope (Figure 10).

There are two possible ways in which to obtain hydrogen:
an exchange route which leads to HD or H2: or a non-
exchange route where no hydride is involved, leading

to D2.

79

 

 

 

 

 

Table 11. Composition of Liberated Gas From DCl Treated
Reduced ® -Cp2TiCl2.
% Isotope
D2 HD H2 Reaction Conditions
A7 35 18 +2A°C BuLi-hexane-Ar
35 ”1.5 23.“ -780 n
A0.7 39.5 19.2 0° n
37.9 39.A 22.7 0° BuLi-hexane-H2
57.0 39.2 A.l 0° i-PngBr-ether-Ar

 

 

80

 

 

 

.oLSpmpooEop coapomop mammo> moouomfi cowopemn pcmomoo mo gamma
00
o¢+ on... 0 our o_¢| o_o.. 0.»:
_ — — — F — _ O
_ _ _ _
lla—
ld’ i®i IlluON
m.
.llon u.
m
m
B o
«a NH
O
n 1...:

.OH spawns

 

81

Exchange route:

p
Q DCI @7/0 Q/H -HD DCI Ti/Cl

Ti h—-D i\\.___4y ‘\.

@ —"@<c. —*@\c, —"’@ c. on

Non-exchange route:

Q" ___’ Qn<LD$® \/°-+_.@Zn\c
o o 0 782a; a

 

 

Consider the rates of reaction with DCl that would be
affected by the temperature of reaction. If the hydride
is formed initially, the exchange process would give
mostly H2 and some D2. These two species would have the
same temperature relationship, i.e., the same slope, for
the graph of percent species ye. temperature. For the
non-exchange process, only HD is formed; this process would
result in a slope differing from that for H2 and D2. If
a hydride is not the initial species, then the exchange
route would give both HD and H2 with the same slope, and
the non-exchange route would lead to formation of D2,

with a different slope. The observed graph fits the latter

82

case, implying a titanocene species that is not initially
a hydride. The B-hydrogen abstraction from a cyclopenta-
dienyl ring by the metal to form a hydride is then a
reasonable explanation of the data. The data suggest
that both titanocene and the hydride obtained by rearrange-
ment are present as the polymer attached species.

A titanocene dihydride which is formed by B-hydrogen

abstraction from the titanocene dialkyl is another pos-

 

 

sibility.
(EECP C><:
p
®'Cp TiCl \ ./°” \ ./3"
2 2 4 /TI\ /TI s‘
(CH

(:AcpzTiCEz p\\ ff” (:}Cp
+ 4h____ r

+———
”2‘1101P02 -Cp//’ FL]

 

Treatment of the homogeneous dihydride with D20 was reported
to give evolution of a gas mixture of H2, HD, and D2

in a ratio of 1:2:1 (Brintzingerso) and l/2:2:2 (Volpin

and Shur58). These values for the isotopic mixture of

gases are reasonably close to the results obtained from

the reaction of the titanocene species beads and DCl,

83

but do not match the theoretical yields for a dihydride.
ESR evidence and observation of reactions indicate that
dihydride formation occurs in ether solvents but not hydro—
carbon solvents. Furthermore, the esr of the reduced
polymer supported species does not match the frozen glass
spectrum of the homogeneous (Cp2TiH2)'. The results for
the isotope mixture do not match the theoretical yields
well enough to presume that (Cp2TiH2)- is the species on
the polymer and other evidence suggests that the dihydride
is not present.

In hexane BuLi treatment of®-Cp2TiC12 might give
36

the monohydride, ®—Cp2Ti-H. Bercaw treated the homo-
geneous monohydride with DC1 (91.5% isotopic purity) to
obtain hydrogen gas of the composition 12% H2, 83% HD,
and 60% D2. The overall equation for the reaction indi-
cates the % D2 should be 1/2 that of the percent HD.

Cp2Ti-H + 2DC1—->Cp2TiCl + HD + 1/202

2
Treatment of the polymer attached reduced species gave
roughly equal amounts of HD and D2 (an approximate average
ratio, H2:HD:D2, was 1:2:2). Results for the bead species
do not correspond well to either the predicted or the actual
isotope yields obtained by Bercaw for (Cp2TiH)2.

The DC1 treatment results indicate that some hydride

intermediate is available, but the identity of the hydride

8A

or the mechanism by which it is formed are as yet un-
known. Ring hydrogen abstraction to form a hydride inter-
mediate seems to fit the data reasonably well. Other
types of evidence, however, are needed to identify the

species on the beads.

ESR of the Copolymer Bead Species

Electron Spin Resonance (esr) spectra of the titano-
cene species attached to the copolymer beads gives some
clues as to the identify of the reduced species. Reduc-
tion of®-Cp2TiCl2 with excess BuLi, NaNp, or RMgBr
leads to gray beads that have an esr signal consisting of
a slightly distorted singlet (Figure 11). The g values
obtained for most samples in the temperature span of +A0°
to -150° are in the range g = 1.990 1 0.002 with an average
signal width of 28 gauss. The lack of detailed hyperfine
splitting precludes a positive identification of the
reduced species on the beads. Since both Ti (II) and Ti

(III) compounds are known that have g values in the range

of g = 1.975 through g 1.998, even the oxidation state of
the titanium cannot be exactly determined.

Either BuLi in hexane or NaNp in THF reduce the titano-
cene dichloride beads to the same species, as indicated by
the esr signal. The g values are the same when reduction

is under argon or under nitrogen at room temperature.

Although LaulLI has reported that a doublet can be obtained

85

.m

 

SEmaoumwmcoosoos Sam co easeooam mmm .2 osswrm
mmsau
8:. .83 coon . SN... 83. 83”
m t. a .. A

 

 

86

by BuLi reduction of very low Ti content beads under a
hydrogen atmosphere, experiments with beads containing
10 times more Ti have not given the multiple peaks ob-
served by Lau. Reductions in pure THF lead to beads with
an esr signal that is more distorted but has a g value in
the same range as before. It would appear that, perhaps,
with the exception of reduction under hydrogen, the same
titanocene species is generated within the polymer beads
under a wide variety of conditions and reducing agents.
Controlled addition of reducing agents can lead to a
stepwise reaction. Addition of one equivalent of NaNp
or i-PngBr to® -Cp2TiC12 gives beads with a A peak
pattern in the esr spectrum, g = 1.979 (Figure 12).
When 1 more equivalent or an excess of reducing agent is
then added, the distorted singlet signal, g = 1.990,
appears. The species with the A peak pattern is probably
®-Cp2TiCl. Homogeneous titanocene monochloride has a
g value of 1.980 which corresponds well to the g value of
the species on the polymer. If the homogeneous Cp2TiCl
solution is frozen at -110°, a distortion of its singlet
occurs and a A peak pattern at g = 1.980 appears (Figure
13). The splitting between peaks matches well the splitt-
ings in the bead esr spectra. Polymer attached Cp2TiCl
can be prepared by reaction of Et2A1Cl with®—Cp2TiC12
and the esr spectrum matches both the homogeneous Cp2TiCl

and the 1 equivalent reduced bead species.

87

3240 3270 3310
I I

 

 

 

Figure 12. ESR spectrum of®-Cp2TiCl.

 

88

3280 3320 ' 3%60
I I

Figure 13. Frozen solution esr spectrum of Cp2TiC1.

89

It is possible then to identify some titanocene deriva-
tives attached to the polymer beads if the esr spectrum
patterns are compared to the patterns of the frozen solu—
tions spectra of the homogeneous compounds. Another com—
pound,®-Cp2Ti(CH2Ph), was identified by this method.

The symmetry and oxidation state of the Cp2Ti(CH2Ph) is
similar to that of Cp2TiCl. Thus, the esr signal of
(:)-Cp2Ti(CH2Ph) is a A peak pattern like that of the polymer
attached monochloride but with different splitting between
peaks and with a g value of 1.983, (Figure 1A). These
values matched the frozen solution spectra for homo-

geneous Cp2Ti(CH2Ph).

A number of homogeneous titanocene derivatives were
prepared and their esr spectra obtained for comparison to
the reduced species on the beads. The temperature, solvent,
and amount of reducing agent added have significant ef—
fects on what reduced titanocene species is obtained.

In ether solvents or benzene at temperatures below —A0°,

Ti (III) dialkyl derivatives are obtained. Brintzingersg
had prepared and recorded esr spectra for a number of these
compounds. At Ti concentration of 10'2M when a large
excess (AD-60 fold) of reducing agent, RLi or RMgBr

(R=Me-, Et-, i-Pr-, Bu—), is added to a cold solution of
CpZTiCl, the titanocene dialkyl anion can be detected by
esr. By BuLi reduction the anion [Cp2Ti(Bu)2]’ was ob-

tained; its esr spectrum consisted of a quintriplet,

 

90

3330 3260 3300
| I

 

 

Fi ure .
g 1A ESR spectrum of®- Cp2Ti(CH2Ph).

91

g = 1.989 (Figure 15). The hyperfine splitting is due
to interaction of the Ti (III) (d1) with the A d-hydrogens
on the two butyl ligands. The number of hyperfine lines
can be determined by the (2n(I)l/2 + 1) rule (I = 1/2).
When the cold solution was warmed slowly an intermediate
[Cp2Ti(Bu)H]' (Figure 16) was obtained which was formed by
B-hydrogen abstraction from a butyl ligand. The esr
spectrum consists of a doublet (7.A G splitting), due to
the Ti (III)-hydride interaction, which is split into
triplets (2.5 G) by the o-hydrogens on the butyl group.
Further warming to room temperature of this solution
yielded [Cp2TiH2]'. The esr spectrum (Figure 17) is a
triplet (7.5 G) due to the two hydride ligands. Under
certain conditions the associated cation can cause further
splitting of the triplet as in Figure 18 where the di-
hydride anion was prepared with i-PngBr; a septet (0.A
G splitting) due to Mg (1 = 3/2) is obtained.

The dihydride anion is stable at room temperature
for several days. Preparative reactions performed at room
temperature yield the dihydride anion rapidly. No dif-
ference in the sequence of esr signals was observed at any
temperature when nitrogen was substituted for argon as the
inert gas during the reaction. Although initial reports59'61
concerning [Cp2T1H2j' attributed nitrogen fixation to the
dihydride anion, subsequent conclusions are that the di-

hydride anion is an inactive side product with regard to

92

3302 3306 33'0

 

 

 

 

 

Figure 15. ESR spectra of Cp2Ti(Bu)2'.

93

32.90 3300 33l0
I I

 

 

 

 

 

 

 

 

 

 

 

 

Figure 16. ESR spectrum of (Cp2Ti(Bu)H-.

9A

3290
3300
I
3310

 

 

 

 

 

 

Figu
re 17
. ES
R spectrum
of
(cp2T1H2)'

95

3293 3296 3300 33.04 3308
I

 

 

 

 

 

 

 

 

 

 

 

 

Figure 18. ESR spectrum of (Cp2TiH2)
splitting.

with Mg hyperfine

96

reaction with nitrogen.u3’62’63 No nitrogen fixation would
be expected then by the reduced polymer beads if the species
on the polymer was K:)-Cp2TiH2]-.

When BuLi in hexane is used to reduce the titanocene
dichloride beads, chemical and esr observations indicate

that a species other than®-Cp2TiH " is being formed.

2
Formation of the dihydride anion is not observed for
homogeneous reactions in hexane. Apparently a more polar
ether solvent is needed to help stabilize the anion.
Frozen homogeneous solutions of the dihydride anion give
esr spectra that are different from the distorted singlet
observed for the reduced species on the polymer (Figure 19).
By contrast, the dialkyl anions, that had detailed hyper-
fine in solution, when frozen were merely broad singlets
(Figure 20). Table 12 compares the esr signals observed
for a number of compounds. Signals containing hyperfine
with splittings lose all detail when the solution is frozen.
Signal width often changes also but the g value remains
nearly the same. The thermal instability of the dialkyl
anions, however, suggests that they are not reasonable
possibilities as the reduced polymer supported species
obtained at ambient temperatures. Thus, neither the alkyl
nor dihydride anions are the polymer supported titanocene
species.

The polymer attached®-Cp2Ti(CH2Ph) was prepared by

treatment of®-Cp2TiCl beads with one equivalent of

97

3290 3300 33l0 3320
I

Figure 19. Frozen solution esr spectrum of
(Cp2TiH2)‘.

98

.cofip:
How 0
so we
Hmoom
m mo uofiospchSU
9mm cm
:0
poomm
m

 

 

 

 

u 0:- Tlllll.
moo.
0.8.- .
o 3..

 

.om ossmfim

99

 

 

 

 

 

 

Table 12. Summary of esr Results for Titanocene Deriva—
tives.

Species g Value Width Split
Cp2TiMe2' septet 1.987 2A.5 A.5
Cp2TiBu2' Quintuplet 1.989 7.0 2.0
Cp2TiEt2' Quintuplet 1.989 --—- 2.0
Cp2TiPr2' Triplet 1.982 6.0 2.3
Cp2TiH2— Triplet 1.993 1A.5 7.5
Cp2T101 Singlet 1.980 —---

Frozen Solution Spectra

Cp2TiMe2- Singlet 1.982 18.0
Cp2TiBu2- Singlet 1.988 7.0
Cp2TiEt2' Singlet 1.986 12.0
Cp2TiPr2' Singlet 1.982 8.0
Cp2TiH2- brd signal 01.993 22.0
Cp2TiCl Pattern 1.980 ----

Beads 5
BuLi +®-CpTiCp012 THF distorted singlet 1.9
i-PngBr +®—0pTiCpCl2 THF distorted singlet 1.9

 

 

100

PhCH2MgBr and an equimolar amount of l,A—dioxane. The
brown beads that were obtained had an esr signal con-
sisting of a four peak pattern, g = 1.983 (Figure 1A).
The splitting between peaks and the g value match those
of the frozen homogeneous solution spectrum for Cp2Ti(CH2Ph).
When the brown beads were treated with nitrogen at -80°
no color change was observed and the esr signal remained
the same. Had nitrogen been taken up by the complex the
esr signal and the bead color should have changed. Since
no nitrogen could be fixed using these beads, the conclu-
sion is that the required dimeric complex with bridging
nitrogen cannot be formed with the titanocene derivative
that is attached to the polymer.

Solutions of the purple hydride, (Cp2TiH)2, in hexane
have an esr signal at g = 1.989 (28 G width). Bercaw36
reported that the solid was diamagnetic, as would be
expected for the dimer complex, but that solutions showed
a weak singlet signal. By analogy to solutions of
(Cp2Ti)2 and C5(CH3)5]Ti, in solution a monomer-dimer
equilibrium appears to exist which would account for the
observed esr signal.

Addition of triphenylphosphine (PPh3) to a solution
Of (Cp2TiH)2 in cold hexane causes symmetrical cleavage
of the dimer to form Cp2Ti(PPh3)H. The esr signal consists
of two doublets with spacings of 9.9 and 23.0 G (Figure 21).

The larger split is due to 31F hyperfine interaction and

101

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l I A

Figure 21. ESR spectrum of Cp2Ti(PPh3)H.

102

the smaller split is due to the hydride. Each peak is
further split by 0.6 G into a quintet. Although no
positive explanation of the quintet can be made, it is
probably due to association of a THF molecule. A peak
at higher field, g = 1.968, corresponds to a decomposition
product obtained in greater yield at higher temperatures.
When THF is added to a hexane solution of (Cp2TiH)2,
a fleeting doublet is observed followed by the 12 line
spectrum shown in Figure 22. The basic pattern is a 1:2:1
triplet (8.5 G splitting) each line of which is split into
a quartet (2.0 G splitting). Olivé and Olive have
observed similar spectra.61 The quartet is probably due
to 7Li (I = 3/2) which is reported to have hyperfine
splitting of approximately 2 G. The rest of the spectrum
is harder to explain when the experimental conditions are
considered. A dihydride would, however, explain the trip-
let splitting. One of the structures below might explain
the spectrum.61 Since the Ti concentration (10-2fl)
is so low in the esr experiments an excess of 2 equivalents
of BuLi was added in the initial reaction to form the purple
hydride dimer. Formation of an anion species may be aided

36

by the more polar THF. Bercaw did not observe this esr
spectrum but BuLi was not present in the solution in his

experiment.

APPENDIX A

ATTEMPTED PREPARATION OF POLYMER SUPPORTED
MOLYBDENOCENE DICHLORIDE THROUGH THE

®-CpMoClu INTERMEDIATE

Since titanocene species were obtainable that were
active catalysts when coordinated to a polymer support,
it was suggested that other compounds that were coor-
dinatively unsaturated and electron deficient might also
prove to be active catalysts, when attached to the co-
polymer beads. Molybdenocene is a sixteen electron system
that has been predicted to undergo carbene like reactions6u
and a variety of basic addition and insertion steps that
are of interest for homogeneous catalysis. Although
chromocene is stable and readily prepared,65 neither
molybdenocene nor tungstenocene have been isolated. In-
ert bis(pentaphenylcyclopentadienyl)molybdenum (II) has

66

been reported and the metallocenes have been postulated

as intermediates in a number of reactions.
Tungstenocene intermediates were postulated in the

insertion reactions of olefin and acetylene complexes

67
2

addition to dienes.68 Thomas reported

and in formation of aryl hydrides and
69

i C
generation of the
metallocenes from the dichloride and Na(Hg) which undergo

addition and insecrion reactions typical of organic

10A

 

105

carbenes. The metallocene intermediates can also be
photogenerated from (05H5)2M(CO)7O or from (05H5)2MH2.71
Thus, polymer supported molybdenocene would be expected

to be highly reactive and possibily useful for catalysis,
Just as was the polymer attached titanocene species.

In a manner similar to titanocene, molybdenocene
polymerizes readily with itself69 to form less reactive
compounds. Attachment to a polystyrene backbone would
prevent molybdenocene dimerization and could lead to a
useful hydrogenation or nitrogen fixation catalyst. When
generated as an intermediate, molybdenocene has been
shown to add across the triple bonds of dinitrogen,
acetylenes, and carbon monoxide.69 Tungstenocene is
capable of inserting into C-H bonds of aromatic hydro-
carbonsél4 and would be expected to deactivate quickly
by reaction with the polystyrene support. Molybdenocene
species generated on the polymer support would be expected
to show greater reactivity because of metal center isola-
tion Just as was achieved with the titanocene species.

A synthesis scheme similar to the preparation of
polymer attached titanocene dichloride was initially
envisioned for attached molybdenocene dichloride (Figure
23). Unlike CpTiCl3 the compound CpMoClu (I) is unstable
and difficult to obtain. The compound J_is prepared in
low yield by reaction of hydrogen chloride with oxo-

or oxochloro- complexes.72 While anhydrous the red

106

 

Figure 23. Preparation scheme for®-Cp2MoC12

 

 

CH2 CH2
L+ MOCIS @ Nan @ /CI
I > MO 7* - MO\
/ \ @i c'
CI CI
CI Cl

Figure 2A. Alternative preparation scheme for®-Cp2MoC12.

107

crystalline J is stable at room temperature, but is very
sensitive to hydrolysis, and either is insoluble or re-
acts with common organic solvents.73’7£4 A communication
outlining a preparation using T1Cp which reports a yield
of A0% J has since appeared.75 Preparation of polymer
supported molybdenocene dichloride via the method in
Figure 23 has not been successful because of the unsuit-
able prOperties of compound J.

An alternate preparative method (Figure 2A) was under—
taken where the compound J need only be an intermediate
attached to the polymer support. Bonds et a1.13 had
reported this method for the attachment of Nb, Mo, and W
to the polymer. Only bead color changes and a C1:Mo
ratio of A:l support the supposition that ®-CpMoClu
(JJJ) formed on the polymer. Bonds offered the equations
below to explain the color change of the beads from black

in benzene to yellow in THF.13

0 H

MCIn + Cp-Li+-—-é-§—9CpM01;Li+ (l)
_ + THF

CpMClnLi -—————+cpMCln_1 + LiCl (2)

The free compound J is reported to be red and paramagnetic
with an esr singlet at g = 1.990, width 52 0.72 Beads
obtained by my preparative method for III do not match the

free compound for those two properties. Reactions (1)

 

108

and (2) seem unlikely in that the polar THF would be more
suitable for stabilizing the polar, charged CpMoCl'S'Li+
than would benzene. There are several other compounds
which could account for the observed data, especially
considering the conditions used for reaction.

The procedure to prepare ®-CpMoClu used was to stir
the®-Cp'Li+ beads in a solution of MoCl5 in benzene under
a dinitrogen atmosphere. Slightly different results were

obtained depending upon the reaction temperature. In-

 

tensely colored, black or brown, paramagnetic beads were
obtained that had C1:Mo ratios of approximately A:1.
Reactions performed at room temperature yield beads that
have an esr singlet at g = 1.96, width 20 G. When the
reaction is performed by reflux extraction of the MoCl5
from a Soxhlet thimble (the procedure also used by Bonds
and Chrandrasekaran76) the yellow-brown beads, after THF
extraction, have multiple esr signals; the predominate
peak was at g = 1.95, width 58 G. Variations in tempera-
ture during preparation obviously led to different results
even though the beads had a similar appearance.

There are several factors that bear against the suc-
cess of the synthesis of JJJ, Molybdenum pentachloride is
only slightly soluble in benzene and is known to form
adducts with most common solvents.77 Adduct formation
is easily observed by the different colors obtained when
MoCl5 is added to a variety of solvents. That it is not

Just the M0015 adducts absorbed in the beads is clearly

109

Table 13. ESR spectral Values for MoClS,

 

 

esr signal

 

 

M0015 in Color g value width
Benzene black/brown 1.9AA 25 G
THF green 1.9AA 12
ether purple 1.9AA 1A
hexane red 1.9AA l2
powdered MoCl5 black 1.956 55
MoCl5 + beads gray 1.955 A“

 

 

indicated by the esr results and the green rather than
yellow color of the THF adduct. Furthermore, hot benzene
reacts with MoCl5 to form insoluble MoClu readily.77’78
Absorption of the tetrachloride by the beads would give a
paramagnetic compound of the right color and C1:Mo ratio

to match the black beads obtained in the attempted synthesis
of compound JJJ. Adduct formation by the tetrachloride77
would then account for the color change in the beads when
extracted with THF.

2MOCI + C6H6-———>2MOC1u + C6H5C1 + HC1

5
MoClu + THF—>MoC1u°(THF)n

Considering the low stability of J reported by Green7zs73

110

and by Shive7u coupled with the fact that J reacts with
most common solvents, formation of a stable polymer sup-
ported species JJJ at reflux temperatures in benzene or
THF is highly unlikely. If JJJ were formed, the beads
should be red after the MoCl5 adduct is washed out; in—
stead the beads remain black. It may be concluded that
JJJ is not obtained by the preparative procedures used,
and that insoluble MoClu is deposited within the polymer.

Molybdenum tetrachloride, its adducts, and oxo-decomposition

 

products then account for the polymer bound species ob-

tained by this method and not compound III.

Experimental

 

Attempted Preparation of (P) -CpMoClu by Use of MoCJ5
Method A: Dry cyclopentadient substituted copolymer
beads (5 g) were suspended in THF for two hours before

treatment with 10.3 mL of 1.7M CH L1 in ether. After two

3
days the red beads were extracted with 3x20 mL of THF
and with 3x20 mL benzene. The beads were suspended in
benzene and the reaction flask fitted with a Soxhlet
extractor charged with 1.8 g MoClS. The benzene was
boiled to extract the MoCl5 into the reaction flask.
Violent bubbling and white fumes were observed as the
hot benzene contacted the MoCl5 in the Soxhlet thimble.

The brown slurry obtained by this procedure was stirred

with the beads for three days. Excess solvent was

111

removed, the beads dried, and the beads extracted with
benzene. Afterwards, the black beads had an esr spectrum
of a distorted singlet, g = 1.950, width A8 G. Further
extraction of the beads with THF caused an immediate color
change to yellow-brown beads. These beads had a complex
esr spectrum with four maJor peaks, the largest being at

g = 1.9A6, 58 G width.

Method B: Into a flask was placed 3 g of cyclopenta-
diene substituted copolymer beads, 20 mL of THF, and a
three-fold excess of CH3Li in ether. After two days of
being stirred, the beads were extracted with 3x20 mL of
THF and with 3x20 mL benzene. Into a 50 mL septum flask
was placed 1.1 g MoC15 and dry benzene at room temperature
added to dissolve the solid. An ink black solution
of MoCl5 in benzene was transferred using a syringe into
the flask containing the red(:}CpLi beads. After the
mixture was stirred a few minutes the black solution
turned a translucent green. More MoCl5 in benzene was
added until the color change to green no longer occurred.
All solvent was removed after three days and the beads
extracted at room temperature with benzene until no color
was observed in the washings. The black beads had a
distorted singlet with g = 1.9A8 (A5 G width) in its esr
spectrum. After treatment with THF at room temperature,

the beads were rust and had an esr singlet at g = 1.958,

19 G width.

 

112

Method C: The procedure above was followed except
that the(:)-CpLi beads were treated with a two-fold excess
of MoCl5 in THF at room temperature. When wet the beads
were a green color but when vacuum dried turned to tan.
The esr was similar to that above and 0.18 mmol of Mo/g

of copolymer was found.

 

APPENDIX B

PREPARATION AND ATTEMPTS TO REDUCE POLYMER ATTACHED

MOLYBDENOCENE DICHLORIDE

Discussion

 

The procedure developed by John Lee15 for attachment
of metallocenes to a polymer support was followed with
minor modifications. Cyclopentadiene substituted beads
were treated with CH3Li and then with molybdenocene di-
chloride. Sigma—pi ring exchange led to attachment (Fig-
ure 25) and treatment with HC1 generates polymer attached
(:)-Cp2MoClZ. The dry beads were air stable, green—brown,
and paramagnetic. Since free Cp2MoC12 is diamagnetic,
either an impurity was present or (®-Cp2MoC12)+ was
formed. The esr spectrum of the bead species (Figure 26)
was similar to the reported frozen glass spectrum of
(Cp2M0C12)+.79’8O Analysis indicated a Mo concentration
of 0.66 mmol/g beads and a Mo:Cl ratio of 1:2.16. Far-
ir spectra of Cp2MoC12 and polymer attached®~Cp2M0012
have been obtained (Table 1A). The spectrum of the ground
bead species is a reasonable match if the molybdenocene
dichloride anion is also present.

Reduction of molybdenocene dichloride with Na/Hg or

BuLi has been reported to give molybdenocene as an

113

 

 

 

Figure 25. Scheme for the Preparation of polymer supported
molybdenocene dichloride.15

 

115

3335 3375 nus 4155

 

Figure 26. ESR spectrum of beads obtained from®-Cp2MoC1-2
synthesis.

116

Table 1A. Far—ir Spectral Data of Cp2MoCl2 Compounds.

 

 

 

® -CpMonCl2 Cp2MoCl2 Literature79
A60 --- ---
A12 A17W A12w
369 392 388
3A2 356 3A6
316 -—- ———
295m 2928 293vs
270m 262s 262s

217w 212s 208s

 

 

117

intermediate. Molybdenocene undergoes carbene like reac-
tions with CO, N2, and HCECH.69 Several attempts were
made to produce the molybdenocene analog on the polymer
from the dichloride. After addition of a reducing agent
to the beads in hexane or THF, the beads were washed and
a second reagent added. The®-Cp2Mo012, after reduction,
was tested for insertion reaction and for catalytic ac-
tivity. Listed below is a summary of the reactions at-

tempted and the results (Table 15). Only two reactions

 

worked: (1) LiBHu treatment produced the dihydride,
®-Cp2MoH2, and (2) allyl benzene was isomerized by Lin
reduced beads.

M. L. H. Green81 has reported a number of unusual
products when Cp2MoH2 is treated with RLi or RMgBr. An
anionic species, (CpZMoH)-M+ (M = Li, MgBr), is produced
that is capable of a wide variety of reactions. These Mo
compounds show reactivity similar to R' in RLi. The 00-
valent Mo-Li bond is formed by alkane elimination in re-
verse of the usual B-hydrogen transfer and elimination
reaction.

The dihydride is not normally a hydrogenation catalyst.
Olefins and acetylenes with activating groups react with
Cp2MoH2 to form stable insertion compounds instead.82
Whereas titanocene dihydride has an empty orbital avail-
able for bonding with the W system of an olefin, the

molybdenocene dihydride has a filled orbital and is a

118

 

 

 

 

mzmzsomvozmoou @ now zmzmzmovzmmzzo

oz III oofippfi>

oz osoxonnfl mme\azwq

coapmNfipoEomH wow ocoucon Hmfiam mme\qzfiq

coomw ooCHmEop oz :mmmz IIII

motto soflaoz so» are ssfleosxzzqu III:

cosponssoeoms oz oz oooma\ooonm.a dzfiq

tempos mm oz oz ososozufl\mm ozsq to whom

Aomovs oz oz zdomozm Hqsm

.s.fi ca Aoovs oz oz 00 ozfiq to sham
.s.H ca Aoova oz oz 00 ozfiq to Room coosw
Ingmo .womon czwmm III ocwxo£\fiqsm maoaoozanmU
coauofinomoo\pospogm coapomom pcowmom ucow< wcHoSUom mofiomom o:

 

 

.mSozmooumofiE mcofiomom assessor... no inseam .3 canoe

 

 

 

119

oz ozouooo Heads defismpm
momoo Roman new ..-..I 825m mandozoomu
women coopw we» Hum IIII coapoSUmp

ofipozoop Qqu 809%
women csoum

 

mH on Aoovs oz oz 00 Hqsm
mH ca owes oz oz zoomozd fiqsm
ozmoos mm oz oz ozoxozuaxmm Hqsm
cofipmNHgoEomH no» econcoo Hmaam ocwxo£\fiq:m
IIII oz ozoncoo. Hagan IIII mmooozoomuv
cofipofipomoo\posooom coapomom pcommom pcow< wcfioSUom mofiooom oz

 

 

.mmozmooImWQpHs mcofipomom ooposoppm no zmmEESm 6H 2nt

120

fairly stable 18 electron system. The ®-Cp2MoH2 was
tested for its ability to undergo insertion reactions or

81 Iso-

to form the anionic complex reported by Green.

merization of allylbenzene was the only reaction that gave

positive results. In all other cases the bead compound

was unreactive. -u
The lack of reactivity of the molybdenocene dichloride F3

and dihydride is certainly puzzling. It could be that the A

molybdenocene intermediate is not generated by the reducing

 

agents used. Another possibility is that the very reactive
molybdenocene reacts with the polymer or reacts to de-
activate in some way soon after it is generated. What—
ever the reaction, the molybdenocene dichloride when
reduced with BuLi or Lin does not produce a catalyst or

an intermediate capable of insertion reactions.

Experimental

 

Preparation of Polymer Attached Chromocene

To 20% crosslinked polystyrene beads containing
pendent cyclopentadiene was added an excess of BuLi and THF.
The red heads were washed 3 x 20 ml with THF by using
syringe techniques. Dissolved CrCl3(THF)3 in THF was
added to the reaction flask containing the beads along
with some powdered zinc. After being refluxed for two

hours, the solution was cooled and excess reagent removed

121

by extraction with THF to yield green beads. These beads
were then treated with excess Nan in THF overnight.
After extraction of the beads with THF, deep red beads
were obtained. The beads were diamagnetic and mildly air
sensitive.

A sample of free chromocene was prepared for compari-

son by reaction of CrCl3 with excess Nan.83 Scarlet red

 

crystals were obtained upon sublimation. A THF solution

of the red crystals was also diamagnetic.

 

Preparation of Polymer Attached Molybdenocene Dichloride

The procedure of John Lee15 was followed with minor
modifications (Figure 25). To 15 g of polymer attached
cyclopentadiene beads (3A mmol) was added A3 m1 of 1.Am
methyl lithium in ether and 75 m1 freshly distilled THF.
This mixture was stirred for three days under nitrogen and
then the excess methyl lithium removed by syringe. The
red beads were washed with THF A x 20 ml and 9.95 g (33.5
mmol) of molybdenocene dichloride was added along with 75
ml of THF. After being stirred for seven days, excess
reagent was removed by syringe and anhydrous HCl bubbled
into the mixture. Uniform green colored beads were ob-
tained after Soxhlet extraction with THF.

Analysis: Mo 0.66 mmol/g C1:Mo ratio 2.16

C1 1.A3 mmol/g
far-1r A12, 295, 270, 217w cm"l

122

Preparation of Polymer Attached Molybdenocene Dihydride

Polymer attached molybdenocene dichloride and an
excess of LiBHu was loaded into a flask in the dry box.
To this mixture, under a nitrogen atmosphere, was added
.dry THF. A vigorous reaction occurred immediately as the
beads become a dirty yellow. Excess LiBHu was removed by
withdrawal of the suspension by a syringe and needle.
After several washings with THF, the beads were used

directly for other reactions.

Preparation of Polymer Attached Dibutanethiol Molybdenocene

Special care in handling techniques was exercised
because of the foul smell of the thiol compounds. All
reactions were carried out under inert gas in enclosed
glassware. Stock solutions were stored in septum capped
flasks and transferred as required by use of a syringe
and needle. All reactions were carried out in a fume
hood and unreacted thiols were treated with an ethanol-
potassium hydroxide solution to decompose them.

Into a 100 mL sidearm flask with a condenser was placed
2.0 g of®~Cp2MoCl2 beads and 9.0 mmol of BuSH (from a
stock solution of 1:1 Et3N:BuSH). Benzene, toluene, or
THF was added (20 mL) and the mixture heated to reflux
and stirred for 1-3 days. The beads were washed with

solvent to remove excess reagent and vacuum dried.

 

 

123

Red-brown beads were obtained that contained 0.11 mmol

 

Mo/g and 0.10 mmol S/g. Analysis of the remaining chloride
on the beads indicates incomplete reaction of the polymer

attached Cp2MoCl2.

 

APPENDIX C

PREPARATION OF POLYMER ATTACHED

GROUP VI CpM(CO)3H COMPOUNDS

Discussion

 

A variety of group VI metal carbonyl compounds were

attached to the polymer support for the following reasons:

 

1. As intermediates to other compounds;

2. To demonstrate the physical separation of compounds
that dimerize readily when homogeneous;

3. To test for catalytic activity;

A. To attach compounds that could easily be
identified by spectral means;

5. To demonstrate the variety of compounds that

could be supported on the polymer.

The compounds CpM(CO)3H (M = Cr, Mo, W) are useful
intermediates easily identified by strong carbonyl

83’8“ and are known reagents for the

stretches in the ir,
hydrogenation of dienes.85 A dimer, (CpM(CO)3)2, forms
readily when the hydride is exposed to air or heated.
The dimer is also the by-product of diene hydrogenation;
the chromium dimer being capable of being reduced by

hydrogen back to the hydride, whereas the molybdenum

12A

125

85,86

and tungsten compounds cannot. These carbonyl

hydrides were thought to fit the criteria above and thus

to be able to take advantage of polymer support.
Synthesis of(:)-CpM(CO)3H (Cr, Mo, W) was achieved by

a modified procedure used for the preparation of the homo—

 

geneous compounds.87 Direct reaction of molybdenum hexa-

carbonyl in THF with the(:)-CpLi beads leads to yellow

 

beads with weak carbonyl peaks in the infrared. Yields

were improved by reaction of (CH3CN)3M(CO)3 with the
88

 

C)-(Xfld.beads in THF under a nitrogen atmosphere.

. ,. . _ wT
f .
h“ "1

Reaction (1)

 

THF HOAc
®-CpLi + M(CO)6 —s® -CpM(00)3‘ —->®—CpM(CO)3H

Reaction (2)

THF HOAC

® -CpLi+(CH3CN)31\/I(Co)3 ——»® -CpM(CO)g ——> —CpM(CO)3H

The reaction is easily observed by color change of the red
C)-&kfld.beads to yellow beads and by the release of carbon
monoxide or acetonirtile.

Formation of the ®-CpM(CO)3H compounds was confirmed
by observation of the carbonyl stretch in the ir. Nor-
mally the polystyrene absorptions obscure the infrared
spectra of attached compounds but a window between 1700-
2100 cm'1 is available for observation of carbonyl

stretching modes. The number of peaks and their position

126

can be used to identify the attached compounds. Although
the metal hydride stretch is also in this region, it is
too weak to be observed for polymer attached species.
Below is a table comparing the observed infrared spectra

for the supported and for the homogeneous compounds.

 

®—CpMo(CO)3H 2025, 1995, 1925—55 CpMo(CO)3H 2030, 19149,

1913 cm’1

®-CpW(CO)3H 1975, 1900 CpW(CO)3H 2020, 1929

 

Since the time these polymer attached compounds were
prepared, a report has appeared89 for the synthesis of
these compounds in dimethyl formamide on 18% crosslinked
polystyrene copolymer. A similar compound has also been
prepared90 by sigma bond attachment, ®-C6Hu-W(CO)3(CD).
This compound has been shown to catalyze olefin metathesis.

The polymer supported Cp2M(CO)3H compounds have greater
stability in air or when heated than the non-attached
compounds. Since the dimer [Cp2M(CO)3]2 would normally
be formed by decomposition of the hydride, the greater
stability of the polymer supported compounds is because
dimerization is prevented. That the dimer is not formed
is demonstrated by bead color and by the supported com-
pound's ir spectrum. When ®-CpMo(CO)H was tested for
hydrogenation of conJugated dienes, no hydrogen addition

could be detected by gas chromatography of the solution.

127
Apparently a mechanism is not available for hydrogena-
tion by use of the monomer hydride under a hydrogen
atmosphere.

Experimental

Preparation 0f(E1-CpW(CO)3H91a92

 

To 0.A g of beads containing ca. 0.A meq of ®-CpH
was added an excess of n-BuLi in hexane and 15 ml of THF.
The beads turn from light yellow to dark red. After being
stirred for 15 minutes, the excess BuLi was washed out by
using two 12 ml portions of THF. Solid tungsten hexa-
carbonyl (0.29 g) was added and the deep red color of
the beads changed to yellow. The mixture was refluxed
for five hours, washed with two 10 m1 portions of THF,
and the beads treated with glaical acetic acid. After
being washed three times with 15 m1 portions of THF,
the beads were vacuum dried. Pale yellow beads were

obtained.

tungsten analysis 0.0A2 mmol/gm beads

infrared ®-CpW(CO)3H 1975, 1900 cm"1

Method B: Improved yields were achieved by using the
soluble (CH3CN)3Mo(CO)3 instead of Mo(00)6 to treat the

beads. Molybdenum hexacarbonyl was refluxed in dry

 

 

128

acetonitrile for five hours and the solvent removed by
evacuation to obtain (CH3CN)3Mo(CO)3.88
To 0.5 g of the beads was added 2.0 ml 1.6M n-BuLi
and 20 ml THF. After being stirred for 30 minutes, the
beads were washed with THF by using syringe techniques.
SOlid (CH3CN)3Mo(CO)3 was dissolved in 20 ml of THF and
the solution added to the flask with the beads. The mix- FE
ture was stirred for 0.5 hour at 50°. The solvent

was then removed, the beads transferred to a Soxhlet

extractor, and the beads extracted well with THF. Yellow

 

F75?! - 3”?“er

beads were obtained.

Infrared ®-CpMo(CO)3H 2025w, 1995s, 1925-55bd cm‘l

Preparation of(E)-CpMo(CO)3C187

The®-CpMo(CO)3H beads were placed in a septum
capped 50 ml flask which was filled with argon. Ten ml
C01” and ten m1 THF were added by syringe. The beads,
which float on top of the carbon tetrachloride layer,
were stirred vigorously to insure contact. A faint red
color developed and remained as the beads were washed

with THF and vacuum dried.

infrared 2010, 1950 and 1915 cm'1
far-infrared 530sh, 560sh, l80bd cm-1

APPENDIX D

PREPARATION OF POLYMER ATTACHED GROUP IV
ARENE METAL TRICARBONYL COMPOUNDS

Discussion

 

The known synthesis procedures93 for ArM(CO)3 (M =
Cr, Mo, W) are very similar to the conditions used in
making CpM(C0)3H. Since benzene rings of the polystyrene
are more numerous than the®-CpH the possibility of ob-
taining a mixture of compounds or only polymer attached
ArM(CO)3 suggested that a direct synthesis should be

attempted. Pittmangu

has reported preparing ®-ArCr(CO)3
on linear and 2% crosslinked polystyrene by using the
direct method of refluxing the polymer with chromium
hexacarbonyl. Recently, Farona95 has reported that
preparation is possible on 2% crosslinked polymer by
direct refluxing with the chromium hexacarbonyl in hep-
tane. The polymer supported species could be used as a
catalyst for Friedel-Crafts reactions. A number of other
catalysis uses have been reported including Friedel-
Crafts reactions,9°’96 hydrogenations,97’98 carbon tetra-

chloride addition to olefins,99 dehydrohalogenations,100

and polymerization of acetylene.101

Direct reaction of molybdenum hexacarbonyl with plain

129

 

130

20% crosslinked polystyrene beads was attempted using THF
or hexane as solvent. After prolonged refluxing, no
apparent reaction had occurred as indicated by bead color
and lack of 0(00) in the infrared. If dimethoxyethane
(DME) or heptane are used as solvents, refluxing for
several hours yields yellow beads that have infrared peaks

at 1970 and 1885 cm‘l.

 

i.r. spectra

 

 

l ..

®-ArMo(co)3 2% beads 1975, 1890 cm‘ a

@ArMo(CO)3 20% beads 1970, 1885
(toluene)Mo(CO)3 homogeneous 1985, 1915

(xylene)Mo(CO)3 homogeneous 1975, 1910

Both 2% and 20% crosslinked beads reacted in this manner
with molybdenum hexacarbonyl, however, the tungsten deri-
vatives could not be obtained even after prolonged reflux-
ing in heptane. If DME is used, which has a higher boil-
ing point, and ethyl cyanide added, light yellow beads
can be obtained after three days of refluxing. The Jm
eJEm formation of (EtCN)3W(CO)3 and higher reaction tempera-
ture allows preparation of the polymer supported arene
tungsten tricarbonyl.

The poor yields obtained when THF or hexane were used
to prepare®—ArM(CO)3 is attributed to the lower solvent

boiling points. Yields are increased by using

131

(EtCN)3M(CO)3 because of its increased solubility, but
the higher reaction temperature is still required. When
the functionalized®-Cp'Li+ beads are used in THF or
hexane the predominate reaction is expected to yield
®-CpM(CO)3H rather than the arene metal tricarbonyl.
Thus either type of compound can be prepared selectively

on the beads by proper choice of reaction conditions.

Experimental

 

Attempted Preparation of Q’)-ArM(CO)3

Method A: When plain 20% crosslinked beads and an
excess of metal hexacarbonyl (M = M0, W) were refluxed in
THF or hexane, no bead color change was noted. Prolonged
refluxing eventually led to a gray solution with black
particles. The gray beads obtained by this method were
extracted with THF or hexane to remove excess hexacarbonyl.
No V(CO) was observed for the beads, but they did contain
a trace of metal. Prolonged heating in the presence
of light was found to decompose the metal hexacarbonyl
in THF solutions to gray insoluble material. Thus

C)-ArM(CO)3 was not prepared under these conditions.

Method B: Plain beads and (EtCN)3M(CO)3 were refluxed
in THF or hexane in the dark. They remain white and have

no V(CO) peaks in the ir.

132

Preparation of (ID -ArMo (CO ) 3

Plain beads (either 2% or 20% crosslinked) were placed
in a flask with excess molybdenum hexacarbonyl and 30 ml
of heptane or DME. Refluxing and stirring for A-8 hours
gave yellow beads. After Soxhlet extraction, the beads
were vacuum dried. The yellow beads slowly darken over

a period of several days in air or upon heating.

infrared @)—ArMo(CO)3 2% crosslinked 1975, 1890 cm-1
spectra: -1
20% crosslinked 1970, 1885 cm

 

The tungsten compound could not be prepared under these
conditions.
Preparation of(P)-ArW(CO)3

Plain beads were mixed with tungsten hexacarbonyl

and 30 ml DME plus 5 ml dry propionitrile added. The

 

mixture was refluxed under nitrogen for three days.
Initially the white tungsten hexacarbonyl dissolved and
a yellow solution formed during the heating. The beads
slowly turned yellow during the reflux period. Yellow
beads were obtained after Soxhlet extraction that had

peaks in the ir at 1995, 1920 cm‘l.

BIBLIOGRAPHY

 

1.
2.

10.

11.

12.

13.

1A.

15.

16.

BIBLIOGRAPHY

J. C. Bailar, Jr., Cat. Rev.-Sci. Eng., JJ, 17 (197A).

 

J. I. Crowley and H. Rappoport, Accts. Chem. Res.,
2. 135 (1976).

 

R. H. Grubbs, Michigan State University, private com-
munication.

A. BJorkman, Chem. and Eng. News, Dec. 15, 1975, p. 5. [L

 

W. H. Lang, A. T. Jurewicz, W. O. Haag, D. Whitehurst, _
and L. D. Rollman, J. Organometal. Chem. , 13A, 85 i

 

 

(1977)

W. Haag and D. D. Whitehurst, Belgian Patent 721,686

(1968). ..
D. C. Neckers A. Kooistra, and G. W. Green, J. Am.

Chem. Soc. , 9A, D9285 (1972).

 

C. U. Pittman, Jr., Chem. Tech., J, A16 (1971).

 

(a) F. R. Hartley and P. N. Vezey, Adv. Organometal.
Chem., _1_7, 189 (1977).

 

 

 

(b) Zofia, M. Michalska, and D. E. Webster, Platinum
Metal Rev., JJ, 65 (197A).

 

W. gaag and D. D. Whitehurst, German Offen. 1,800,379

R. H. Grubbs and L. C. Kroll, J. Am. Chem. Soc., 2_
3062 (1971)

 

C. U. Pittman, Jr. , and G. 0. Evans, Chem. Tech. 3,
560 (1973)

W. D. Bonds, Jr., C. H. Brubaker, Jr., E. S. Chandrase-
karan, C. Gibbons, R. H. Grubbs, and L. C. Kroll, J.
Amer. Chem. Soc., 21, 2128 (1975).

 

C. P. Lau, Thesis, Michigan State University, 1977.

J. G. Lee and C. H. Brubaker, Jr., J. Organometal Chem. ,
135,115 (1977)

"The Federal Register", Vol. 39, No. 20, Part III,
Occupational Safety and Health Administration, "Carcino-
gens".

 

133

13A

17. (a) A. Davison and S. S. Wreford, J, gm. Chem. Soc.,
‘29, 3017 (197“)-

(b) H. H. Brintzinger and J. E. Bercaw, J. Am. Chem.
Soc., 92, 6182 (1970).
18. G. W. Watt, L. J. Baye, and F. 0. Drummond, Jr.,
J. A_. Chem. Soc., 88, 1138 (1966).

 

19. C. L. Gibbons, Thesis, Michigan State University,
1973.

20. C. U. Pittman, Jr., and R. M. Hanes, University of
Alabama, Private Communication.

 

21. K. W. Pepper, H. M. Parsley, and M. A. Young, J.
Chem. Soc., A097 (1953)-

 

 

22. D. A. Skoog and D. M. West, "Fundamentals of Analytical
Chemistry", 2nd Edition, Holt, Rinehart, and Winston,
New York, NY, 1963, p. 233.

23. R. D. Gorsich, J, gm, Chem. Soc., 88, A7AA (1960).

2A. K. Claus and H. Bestian, Justus Liebigs Ann. Chem.,
65A, 8 (1962).

 

25. R. S. P. Coutts, P. C. Wailes, and R. L. Martin, J.
Organomet. Chem., 81, 375 (1973).

 

26. M. L. H. Green and C. R. Lucas, J. Chem. Soc., Dalton
Trans., 1000 (1972).

 

 

27. J. H. Teuben and H. J. De Liefde Meijer, J. Organo-
metal. Chem., 88, 313 (1972).

 

28. F. W. Van Der WeiJ and J. H. Teuben, J. Organometal.
Chem., 105, 203 (1976).

 

29. D. G. Breslow and N. R. Newburg, g. 58. Chem. Soc.,
91. 81 (1959).

30. J. H. Teuben, J. Organometal. Chem., 81, 159 (1973).

 

31. M. E. Volpin and V. B. Shur, "Organometallic Reactions",
Vol. I, E. I. Becker and M. Tsutsui, Eds., Wiley-
Interscience, New York, NY, 1970, p. 102.

32. L. Kroll Thesis, Michigan State University, 1976.

33.

3M.

35.

36.
37.

38.

39.

MO.

41.

A2.
”3.

uu.

“5.

H6.

“7.
U8.

135

(a) H. H. Brintzinger, J. Am. Chem. S22., 82, 6871
(1967).

 

(b) G. H. Olive' and S. Olive' Angew Chem. Int.
gg., g, 650 (1969). ’ ’ ’

 

M E. Volpin, V. A. Dubovitskii, O. V. Nogina, and
N. Kursanov, Dokl. Akad. Nauk SSSR, 151, 1100 (1963).

 

 

D.
H. H. Brintzinger and L. S. Bartell, J. Am, Chem. Soc.,

J. E. Bercaw, Thesis, University of Michigan, 1971.

I. N. Ivleva, A. K. Shilova, S. I. Salienko, and Y. G.
Borad'ko, Dokl. Akad. Nauk. SSSR, 213, 116 (1973).

 

P. C. Nailes, R. S. P. Coutts, and H. Weigold,"0rgano-
metallic Chemistry of Ti, Zr and Hf",Academic Press,
Inc., New York, NY, 1974, p. 249.

E. E. van Tamelen, R. B. Fechter, S. w. Schneller, G.
Boche, R. H. Greeley, and B. Akermark, J. Am. Chem.

Soc., 91, 1551 (1969).

M. E. Volpin and M. A. Ilatovskaya, L. V. Kosyakova,
and V. B. Shur, J. Chem. Soc., Chem. Comm., 107“
(1968). -

G. P. Pez, and S. C. Kwan, J. Am. Chem. Soc., 98,
8079 (1976).

A. E. Shilov, Russian Chem. Rev., 53, 378 (197A).

 

 

 

 

J. E. Bercaw, et a1., J. Am. Chem. S93., 98, 8358
(1976).

M. E. Volpin, V. A. Duovitskii, O. V. Nogina, and D.
N. Kursanov, Dokl, Akad. Nauk. SSSR, 151, 1100 (1963).

 

W. Kirmse,"Carbene Chemistry",Academic Press, New
York, NY, 196“.

Y. G. Borod'ko, A. E. Shilov, and A. A. Shteinman,
Dokl, Akad. Nauk. SSSR, 168, 581 (1966).

 

 

 

R. H. Marvich, Thesis, University of Michigan, 1971.

G. J. Olthof, J. Organometal. Chem., 128, 367 (1977).

 

“9.

50.
51.

52.

53.

5A.
55.

56.

57.

58.

59.
60.

61.

62.
63.

6H.

65.

66.

67.

136

J. E. Bercaw, R. H. Marvich, L. G. Bell, and H. H.
Brintzinger, 9. 52' Chem. Soc., 99. 1219 (1972).

H. H. Brintzinger, 9. 9m. Chem. 923., Q9, 6871 (1967).

 

M. E. Volpin, V. B. Shur, 33 gl., Dokl, Akad. Nauk.
SSSR. 19“, 577 (1970).

 

J. S. Kenworthy, J. Myatt, and M. C. R. Symons, 1.
Chem. Soc. (A), 1020 (1971).

 

J. X. McDermott, M. E. Wilson, and G. M. Whitesides,
J. Am. Chem. Soc., 9Q, 6529 (1976).

M. H. Peng, Thesis, Michigan State University, 1977.

 

M. D. Rausch, W. H. Boon and H. G. Alt, J. Organometal.
Chem., 1M1, 299 (1977L

 

 

Y.G. Borodko, I. N. Ivleva, L. M. Kachapina, S. I.
Salienko, A. K. Shilova, and 0. E. Shilov, 9. Chem.
Soc. Comm., 1178 (1972).

 

J. H. Teuben, g. Organometal. Chem., g9, 291 (197A).

M. E. Volpin, V. B. Shur, et a1. , g. Organometal.
Chem. 27, C5 (1971)

 

 

H. H. Brintzinger, g. Am. Chem. Sgg.,‘§§, H307 (1966).

 

V. B. Panov, Y. M. Shul'ga, E. F. Kvashina, and Y. G.
Borod'ko, Kinetika i. Kataliz., 99, 518 (197U).

 

G. H. Olive and S. Olive, Angew. Chem., Int. 99., l
386 (1968). -—-

 

E. Bayer and V. Schurig, Chem. Ber., 102, 3378 (1969).

 

C. Ungnienasu and E. Streba, J. Ino_g. Nucl. Chem. ,
L3, 3753 (1972).

K. L. T. Wong and H. H. Brintzinger, 9. Am. Chem. Soc.
97, 5193 (1975)-

G. Wilkinson, F. A. Cotton, and J. M. Birmingham, 9.
Inorg. Nucl. Chem., g, 95 (1956).

 

 

 

W. Hubel and R. Mereny1,J. 0rganometa1.,Chem., 9,
213 (196AL

 

B. R. Francis, M. L. H. Green, and G. G. Roberts,

Chem. Commun., 1290 (1971).

137

68. M. L. H. Green and P. R. Knowles, J. Chem. Soc.,
1508 (1971).

69. J. L. Thomas, J. Am. Chem. Soc., 95, 1838 (1973).

 

70. K. L. T. Wong, J. L. Thomas, and H. H. Brintzinger, J.
Am. Chem. Soc., 96, 369A (197A).

 

71. G. L. Geoffroy and M. G. Bradley, J. Organometal Chem.,
13“, 027 (1977).

72. M. Cousins and M. L. H. Green, J. Am. Chem. Soc., 1567
(196“)-

73. M. Cousins and M. L. H. Green, J. Am. Chem. Soc.,
16 (1969).

 

 

74. L. Shive, Michigan State University, private communica-
tion, 1977.

75. M. J. Bunker, A. DeCian, and M. L. H. Green, Chem.
Commun., 59 (1977)-

76. W. D. Bonds and E. S. Chandrasekaran, Michigan State
University, Private communication, 1973.

77. M. L. Larson and F. W. Moore, Inorg. Chem., 3, 285
(196“).

 

78. T. M. Brown and E. L. McCann, III, Inorg. Chem., 1,
1227 (1968).

 

79. R. L. Cooper and M. L. H. Green, J. Am. Chem. Soc.,
1155 (1967).

80. w. M. E. Lindsell, J, Chem. §22., (D), 25u8 (1975).

 

81. F. W. Benfield, B. R. Francis, M. L. H. Green. N.
T. Luoug-Thi, G. Moser, J. S. Poland, and D. M. Roe,
J. Less-Comm. Metals., 36, 187 (197H).

 

82. A. Nakamura and S. Otsuka, J. Am. Chem. Soc., 25,
1886 (1972).

 

83. J. J. Eisch and R. B. King,"0rganometallic Synthesis,"
Vol. 1., Academic Press, New York, NY 1965.

8A. F. A. Cotton and G. Wilkinson,"Advanced Inorganic
Chemistry",Interscience Publishers, New York, NY,

1972.

85. A. Miyake and H. Kondo, Angew. Chem., Int. EJ., l,
880 (1968).

86.

87.

88.

89.

90.

91.

92.

93.

9M.

95.

96.

97.

98.

99.

100.

101.

12:, 2g.. g, 520 (1969).

19, P5 (1969).

2% , 9_0, was (1968).

138

A. Miyake, H. Kondo, and M. Aoyama, Angew, Chem.,

 

T. S. Piper and G. Wilkinson, J. Inorg. Nuc1.Chem.,
3, 10h (1956L

R. 26 King and A. Fronzaglia, Inorg. Chem. , 5, 1837
19 L ’

 

 

G. Gubitosa and H. H. Brintzinger, J. Organometal.
Chem., 1&0, 187 (1977).

 

S. Warwel and P. Bushmeyer, Angew. Chem., Int. E§°

D. P. Tate, W. R. Knipple, and J. M. Angel, Inorg.

 

Chem., J, “33 (1962).

S. A. Keppie and M. F. Lappert, J. Organometal. Chem.

 

B. Nicholls and M. C. Whiting, J. Chem. Soc. , 51
(1951L

(a) C. U. Pittman, et a1. , J. Poly. Sci., J9, 379
(1972)

(b) C. U. Pittman, J. Organometal. Chem., 61, 295
(197“)

 

 

 

C. P. Tsonis and M. F. Farnoa, J. Organometal. Chem. ,
11a, 293 (1976L

M. F. Farona and J. F. White, J. Am. Chem. Soc. , _J,
2826 (1971).

 

E. N. Frankel, E. Selke, and C. A. Glass, J. Am. Chem.

 

C. U. Pittman, Jr., B. T. Kim, and W. M. Douglas, J.
Org. Chem., 39, 590 (1975).

 

O. Gandolfi and M. Cais, J. Organometa1.Chem., 125,
1H1 (1977)

J. F. White and M. F. Farona, J. Organometal. Chem. ,
63. 329 (1973)

P. S. Woon and M. F. Farona, J. Poly. Sci., 12,17h9
(197“)

 

 

 

 

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

umunummulnwmm

3 3175 8